Re: Location of goal/purpose was Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Will, --- On Tue, 7/15/08, William Pearson [EMAIL PROTECTED] wrote: And I would also say of evolved systems. My fingers purpose could equally well be said to be for picking ticks out of the hair of my kin or for touch typing. E.g. why do I keep my fingernails short, so that they do not impede my typing. The purpose of gut bacteria is to help me digest my food. The purpose of part of my brain is to do differentiation of functions, because I have . Actually, I agree with that, good point. No matter what kind of system, designed or evolved, it has no intrinsic purpose, only a purpose we interpret. Purpose in other words is a property of the observer, not the observed. If you want to think of a good analogy for how emergent I want the system to be. Imagine someone came along to one of your life simulations and interfered with the simulation to give some more food to some of the entities that he liked the look of. This wouldn't be anything so crude as to specify the fitness or artificial breeding, but it would tilt the scales in the favour of entities that he liked all else being equal. Would this invalidate the whole simulation because he interfered and bought some of his purpose into it? If so, I don't see why. No, it certainly wouldn't invalidate it. That is in fact what I would do to nudge the simulation along, provide it with incentives for developing in complexity, adding richness to the environment, creating problems to be solved. So unless you believe that life was designed by God (in which case the purpose of life would lie in the mind of God), the purpose of the system is indeed intrinsic to the system itself. I think I would still say it didn't have a purpose. If I get your meaning right. Will Yes, that's what I would say (now). Here's the clearest way I can put it: purpose is a property of the observer - we interpret purpose in an observed system, and different observers can have different interpretations. However, we can sometimes talk about purpose in an objective sense in the observed system, *as if* it had an objective purpose, but only to the extent that we can relate it to the observed goals and behavior of the system (which, ultimately, are also interpreted). Which is another way of showing that when we examine concepts like goals, purpose, and behavior, we ultimately come back to the fact that these are mental constructions. They are our maps, not the territory. Terren --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=108809214-a0d121 Powered by Listbox: http://www.listbox.com
Re: Location of goal/purpose was Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
2008/7/14 Terren Suydam [EMAIL PROTECTED]: Will, --- On Fri, 7/11/08, William Pearson [EMAIL PROTECTED] wrote: Purpose and goal are not intrinsic to systems. I agree this is true with designed systems. And I would also say of evolved systems. My fingers purpose could equally well be said to be for picking ticks out of the hair of my kin or for touch typing. E.g. why do I keep my fingernails short, so that they do not impede my typing. The purpose of gut bacteria is to help me digest my food. The purpose of part of my brain is to do differentiation of functions, because I have . The designed system is ultimately an extension of the designer's mind, wherein lies the purpose. Oddly enough that is what I want the system to be. Rather an extension of my brain. Of course, as you note, the system in question can serve multiple purposes, each of which lies in the mind of some other observer. The same is true of your system, even though its behavior may evolve. Your button is what tethers its purpose to your mind. On the other hand, we can create simulations in which purpose is truly emergent. To support emergence our design must support large-scale, (global) interactions of locally specified entities. Conway's Game of Life is an example of such a system - what is its purpose? To provide an interesting system for researchers to research cellular automata? ;) I think I can see your point, It has no practical purpose as such. Just a research purpose. It certainly wasn't specified. And neither am I specifying the purpose of mine! I'm quite happy to hook up the button to something I press when I feel like it. I could decide the purpose of the system was to learn and be good at backgammon one day, in which case my presses would reflect that, or I could decide the purpose of the system was to search the web. If you want to think of a good analogy for how emergent I want the system to be. Imagine someone came along to one of your life simulations and interfered with the simulation to give some more food to some of the entities that he liked the look of. This wouldn't be anything so crude as to specify the fitness or artificial breeding, but it would tilt the scales in the favour of entities that he liked all else being equal. Would this invalidate the whole simulation because he interfered and bought some of his purpose into it? If so, I don't see why. The simplest answer is probably that it has none. But what if our design of the local level was a little more interesting, such that at the global level, we would eventually see self-sustaining entities that reproduced, competed for resources, evolved, etc, and became more complex over a large number of iterations? Then the system itself still wouldn't have a practical purpose. For a system Y to have a purpose, you have to have be able to say part X is like it is for Y to perform its function. Internal state corresponding to the entities might be said to have purpose, but not the system as a whole. Whether that's possible is another matter, but assuming for the moment it was, the purpose of that system could be defined in roughly the same way as trying to define the purpose of life itself. We have to be careful here. What meaning of the word life are you using? 1) The biosphere + evolution 2) And individuals exsistance. The first has no purpose. You can never look at the biosphere and figure out what bits are for what in the grander scheme of things, or ask yourself what mutations are likely to be thrown up to better achieve its goal. That we have some self-regulation on the Gaian scale is purely anthropic, biospheres without it would likely have driven themselves to a state not able to support lives. An individual entity has a purpose, though. So to that extent the purposeless can create the purposeful. So unless you believe that life was designed by God (in which case the purpose of life would lie in the mind of God), the purpose of the system is indeed intrinsic to the system itself. I think I would still say it didn't have a purpose. If I get your meaning right. Will --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=108809214-a0d121 Powered by Listbox: http://www.listbox.com
Re: Location of goal/purpose was Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Will, --- On Fri, 7/11/08, William Pearson [EMAIL PROTECTED] wrote: Purpose and goal are not intrinsic to systems. I agree this is true with designed systems. The designed system is ultimately an extension of the designer's mind, wherein lies the purpose. Of course, as you note, the system in question can serve multiple purposes, each of which lies in the mind of some other observer. The same is true of your system, even though its behavior may evolve. Your button is what tethers its purpose to your mind. On the other hand, we can create simulations in which purpose is truly emergent. To support emergence our design must support large-scale, (global) interactions of locally specified entities. Conway's Game of Life is an example of such a system - what is its purpose? It certainly wasn't specified. The simplest answer is probably that it has none. But what if our design of the local level was a little more interesting, such that at the global level, we would eventually see self-sustaining entities that reproduced, competed for resources, evolved, etc, and became more complex over a large number of iterations? Whether that's possible is another matter, but assuming for the moment it was, the purpose of that system could be defined in roughly the same way as trying to define the purpose of life itself. So unless you believe that life was designed by God (in which case the purpose of life would lie in the mind of God), the purpose of the system is indeed intrinsic to the system itself. Terren --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=108809214-a0d121 Powered by Listbox: http://www.listbox.com
Re: Formal proved code change vs experimental was Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
William, On 7/7/08, William Pearson [EMAIL PROTECTED] wrote: 2008/7/3 Steve Richfield [EMAIL PROTECTED]: William and Vladimir, IMHO this discussion is based entirely on the absence of any sort of interface spec. Such a spec is absolutely necessary for a large AGI project to ever succeed, and such a spec could (hopefully) be wrung out to at least avoid the worst of the potential traps. And if you want the interface to be upgradeable, or alterable what then? This conversation was based on the ability to change as much of the functional and learning parts of the systems as possible. You should read the X.25 (original US version) or EDIFACT(newer/better European version) EDI (Electronic Data Interchange) spec. There are several free downloadable EDIFACT descriptions on-line, but the X.25 people want to charge for EVERYTHING. This is the basis for most of the world's financial systems. It is designed for smooth upgrading, even though some users on a network do NOT have the latest spec or software. The specifics of various presently defined message types aren't interesting in this context. However, the way that they make highly complex networks gradually upgradable IS interesting and I believe provides a usable roadmap for AGI development. When looking at this, think of this as a prospective standard for RPC (Remote Procedure Calls). Steve Richfield --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: Formal proved code change vs experimental was Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
2008/7/3 Steve Richfield [EMAIL PROTECTED]: William and Vladimir, IMHO this discussion is based entirely on the absence of any sort of interface spec. Such a spec is absolutely necessary for a large AGI project to ever succeed, and such a spec could (hopefully) be wrung out to at least avoid the worst of the potential traps. And if you want the interface to be upgradeable, or alterable what then? This conversation was based on the ability to change as much of the functional and learning parts of the systems as possible. Will Pearson --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Terren, Remember when I said that a purpose is not the same thing as a goal? The purpose that the system might be said to have embedded is attempting to maximise a certain signal. This purpose presupposes no ontology. The fact that this signal is attached to a human means the system as a whole might form the goal to try and please the human. Or depending on what the human does it might develop other goals. Goals are not the same as purposes. Goals require the intentional stance, purposes the design. To the extent that purpose is not related to goals, it is a meaningless term. In what possible sense is it worthwhile to talk about purpose if it doesn't somehow impact what an intelligent actually does? Does the following make sense? The purpose embedded within the system will be try and make the system not decrease in its ability to receive some abstract number. The way I connect up the abstract number to the real world will the govern what goals the system will likely develop (along with the initial programming). That is there is some connection, but it is tenuous and I don't have to specify an ontology. Will --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Will, --- On Fri, 7/4/08, William Pearson [EMAIL PROTECTED] wrote: Does the following make sense? The purpose embedded within the system will be try and make the system not decrease in its ability to receive some abstract number. The way I connect up the abstract number to the real world will the govern what goals the system will likely develop (along with the initial programming). That is there is some connection, but it is tenuous and I don't have to specify an ontology. Will I don't think I follow, but if I do, you're saying that the purpose of your system determines the goals of the system, which sounds like it's just semantics... Terren --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
2008/7/3 Terren Suydam [EMAIL PROTECTED]: --- On Wed, 7/2/08, William Pearson [EMAIL PROTECTED] wrote: Evolution! I'm not saying your way can't work, just saying why I short cut where I do. Note a thing has a purpose if it is useful to apply the design stance* to it. There are two things to differentiate between, having a purpose and having some feedback of a purpose built in to the system. I don't believe evolution has a purpose. See Hod Lipson's TED talk for an intriguing experiment in which replication is an inevitable outcome for a system of building blocks explicitly set up in a random fashion. In other words, purpose is emergent and ultimately in the mind of the beholder. See this article for an interesting take that increasing complexity is a property of our laws of thermodynamics for non-equilibrium systems: http://biology.plosjournals.org/perlserv/?request=get-documentdoi=10.1371/journal.pbio.0050142ct=1 In other words, Darwinian evolution is a special case of a more basic kind of selection based on the laws of physics. This would deprive evolution of any notion of purpose. Evolution doesn't have a purpose, it creates things with purpose. Where purpose means it is useful to apply the design stance on it, e.g. ask what an eye on a frog is for. It is the second I meant, I should have been more specific. That is to apply the intentional stance to something successfully, I think a sense of its own purpose is needed to be embedded in that entity (this may only be a very crude approximation to the purpose we might assign something looking from an evolution eye view). Specifying a system's goals is limiting in the sense that we don't force the agent to construct its own goals based on it own constructions. In other words, this is just a different way of creating an ontology. It narrows the domain of applicability. That may be exactly what you want to do, but for AGI researchers, it is a mistake. Remember when I said that a purpose is not the same thing as a goal? The purpose that the system might be said to have embedded is attempting to maximise a certain signal. This purpose presupposes no ontology. The fact that this signal is attached to a human means the system as a whole might form the goal to try and please the human. Or depending on what the human does it might develop other goals. Goals are not the same as purposes. Goals require the intentional stance, purposes the design. Also your way we will end up with entities that may not be useful to us, which I think of as a negative for a long costly research program. Will Usefulness, again, is in the eye of the beholder. What appears not useful today may be absolutely critical to an evolved descendant. This is a popular explanation for how diversity emerges in nature, that a virus or bacteria does some kind of horizontal transfer of its genes into a host genome, and that gene becomes the basis for a future adaptation. When William Burroughs said language is a virus, he may have been more correct than he knew. :-] Possibly, but it will be another huge research topic to actually talk to the things that evolve in the artificial universe, as they will share very little background knowledge or ontology with us. I wish you luck and will be interested to see where you go but the alife route is just to slow and resource intensive for my liking. Will --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
2008/7/2 Vladimir Nesov [EMAIL PROTECTED]: On Thu, Jul 3, 2008 at 12:59 AM, William Pearson [EMAIL PROTECTED] wrote: 2008/7/2 Vladimir Nesov [EMAIL PROTECTED]: On Wed, Jul 2, 2008 at 9:09 PM, William Pearson [EMAIL PROTECTED] wrote: They would get less credit from the human supervisor. Let me expand on what I meant about the economic competition. Let us say vmprogram A makes a copy of itself, called A', with some purposeful tweaks, trying to make itself more efficient. So, this process performs optimization, A has a goal that it tries to express in form of A'. What is the problem with the algorithm that A uses? If this algorithm is stupid (in a technical sense), A' is worse than A and we can detect that. But this means that in fact, A' doesn't do its job and all the search pressure comes from program B that ranks the performance of A or A'. This generate-blindly-or-even-stupidly-and-check is a very inefficient algorithm. If, on the other hand, A happens to be a good program, then A' has a good change of being better than A, and anyway A has some understanding of what 'better' means, then what is the role of B? B adds almost no additional pressure, almost everything is done by A. How do you distribute the optimization pressure between generating programs (A) and checking programs (B)? Why do you need to do that at all, what is the benefit of generating and checking separately, compared to reliably generating from the same point (A alone)? If generation is not reliable enough, it probably won't be useful as optimization pressure anyway. The point of A and A' is that A', if better, may one day completely replace A. What is very good? Is 1 in 100 chances of making a mistake when generating its successor very good? If you want A' to be able to replace A, that is only 100 generations before you have made a bad mistake, and then where do you go? You have a bugged program and nothing to act as a watchdog. Also if A' is better than time A at time t, there is no guarantee that it will stay that way. Changes in the environment might favour one optimisation over another. If they both do things well, but different things then both A and A' might survive in different niches. I suggest you read ( http://sl4.org/wiki/KnowabilityOfFAI ) If your program is a faulty optimizer that can't pump the reliability out of its optimization, you are doomed. I assume you argue that you don't want to include B in A, because a descendant of A may start to fail unexpectedly. Nope. I don't include B in A because if A' is faulty it can cause problems to whatever is in the same vmprogram as it, by overwriting memory locations. A' being a separate vmprogram means it is insulated from the B and A, and can only have limited impact on them. I don't get what your obsession is with having things all be in one program is anyway. Why is that better? I'll read knowability of FAI again, but I have read it before and I don't think it will enlighten me. I'll come back to the rest of your email once I have done that. Will Pearson --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
On Thu, Jul 3, 2008 at 10:45 AM, William Pearson [EMAIL PROTECTED] wrote: Nope. I don't include B in A because if A' is faulty it can cause problems to whatever is in the same vmprogram as it, by overwriting memory locations. A' being a separate vmprogram means it is insulated from the B and A, and can only have limited impact on them. Why does it need to be THIS faulty? If there is a known method to prevent such faultiness, it can be reliably implemented in A, so that all its descendants keep it, unless they are fairly sure it's not needed anymore or there is a better alternative. I don't get what your obsession is with having things all be in one program is anyway. Why is that better? I'll read knowability of FAI again, but I have read it before and I don't think it will enlighten me. I'll come back to the rest of your email once I have done that. It's not necessarily better, but I'm trying to make explicit in what sense is it worse, that is what is the contribution of your framework to the overall problem, if virtually the same thing can be done without it. -- Vladimir Nesov [EMAIL PROTECTED] http://causalityrelay.wordpress.com/ --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
2008/7/3 Vladimir Nesov [EMAIL PROTECTED]: On Thu, Jul 3, 2008 at 10:45 AM, William Pearson [EMAIL PROTECTED] wrote: Nope. I don't include B in A because if A' is faulty it can cause problems to whatever is in the same vmprogram as it, by overwriting memory locations. A' being a separate vmprogram means it is insulated from the B and A, and can only have limited impact on them. Why does it need to be THIS faulty? If there is a known method to prevent such faultiness, it can be reliably implemented in A, so that all its descendants keep it, unless they are fairly sure it's not needed anymore or there is a better alternative. Because it is dealing with powerful stuff, when it gets it wrong it goes wrong powerfully. You could lock the experimental code away in a sand box inside A, but then it would be a separate program just one inside A, but it might not be able to interact with programs in a way that it can do its job. There are two grades of faultiness. frequency and severity. You cannot predict the severity of faults of arbitrary programs (and accepting arbitrary programs from the outside world is something I want the system to be able to do, after vetting etc). I don't get what your obsession is with having things all be in one program is anyway. Why is that better? I'll read knowability of FAI again, but I have read it before and I don't think it will enlighten me. I'll come back to the rest of your email once I have done that. It's not necessarily better, but I'm trying to make explicit in what sense is it worse, that is what is the contribution of your framework to the overall problem, if virtually the same thing can be done without it. I'm not sure why you see this distinction as being important though. I call the vmprograms separate because they have some protection around them, but you could see them as all one big program if you wanted. The instructions don't care whether we call the whole set of operations a program or not. This, from one point of view, is true at least while it is being simulated the whole VM is one program inside a larger system. Will Pearson --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
On Thu, Jul 3, 2008 at 4:05 PM, William Pearson [EMAIL PROTECTED] wrote: 2008/7/3 Vladimir Nesov [EMAIL PROTECTED]: On Thu, Jul 3, 2008 at 10:45 AM, William Pearson [EMAIL PROTECTED] wrote: Nope. I don't include B in A because if A' is faulty it can cause problems to whatever is in the same vmprogram as it, by overwriting memory locations. A' being a separate vmprogram means it is insulated from the B and A, and can only have limited impact on them. Why does it need to be THIS faulty? If there is a known method to prevent such faultiness, it can be reliably implemented in A, so that all its descendants keep it, unless they are fairly sure it's not needed anymore or there is a better alternative. Because it is dealing with powerful stuff, when it gets it wrong it goes wrong powerfully. You could lock the experimental code away in a sand box inside A, but then it would be a separate program just one inside A, but it might not be able to interact with programs in a way that it can do its job. There are two grades of faultiness. frequency and severity. You cannot predict the severity of faults of arbitrary programs (and accepting arbitrary programs from the outside world is something I want the system to be able to do, after vetting etc). You can't prove any interesting thing about an arbitrary program. It can behave like a Friendly AI before February 25, 2317, and like a Giant Cheesecake AI after that. -- Vladimir Nesov [EMAIL PROTECTED] http://causalityrelay.wordpress.com/ --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Formal proved code change vs experimental was Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Sorry about the long thread jack 2008/7/3 Vladimir Nesov [EMAIL PROTECTED]: On Thu, Jul 3, 2008 at 4:05 PM, William Pearson [EMAIL PROTECTED] wrote: Because it is dealing with powerful stuff, when it gets it wrong it goes wrong powerfully. You could lock the experimental code away in a sand box inside A, but then it would be a separate program just one inside A, but it might not be able to interact with programs in a way that it can do its job. There are two grades of faultiness. frequency and severity. You cannot predict the severity of faults of arbitrary programs (and accepting arbitrary programs from the outside world is something I want the system to be able to do, after vetting etc). You can't prove any interesting thing about an arbitrary program. It can behave like a Friendly AI before February 25, 2317, and like a Giant Cheesecake AI after that. Whoever said you could? The whole system is designed around the ability to take in or create arbitrary code, give it only minimal access to other programs that it can earn and lock it out from that ability when it does something bad. By arbitrary code I don't mean random, I mean stuff that has not formally been proven to have the properties you want. Formal proof is too high a burden to place on things that you want to win. You might not have the right axioms to prove the changes you want are right. Instead you can see the internals of the system as a form of continuous experiments. B is always testing a property of A or A', if at any time it stops having the property that B looks for then B flags it as buggy. I know this doesn't have the properties you would look for in a friendly AI set to dominate the world. But I think it is similar to the way humans work, and will be as chaotic and hard to grok as our neural structure. So as likely as humans are to explode intelligently. Will Pearson --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: Formal proved code change vs experimental was Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
William and Vladimir, IMHO this discussion is based entirely on the absence of any sort of interface spec. Such a spec is absolutely necessary for a large AGI project to ever succeed, and such a spec could (hopefully) be wrung out to at least avoid the worst of the potential traps. For example: Suppose that new tasks stated the maximum CPU resources needed to complete. Then, exceeding that would be cause for abnormal termination. Of course, this doesn't cover logical failure. More advanced example: Suppose that tasks provided a chain of consciousness log as they execute, and a monitor watches that chain of consciousness to see that new entries are repeatedly made, that they are grammatically (machine grammar) correct, and verifies anything that is easily verifiable. Even more advanced example: Suppose that a new pseudo-machine were proposed, whose fundamental code consisted of reasonable operations in the logic-domain being exploited by the AGI. The interpreter for this pseudo-machine could then employ countless internal checks as it operated, and quickly determine when things went wrong. Does anyone out there have something, anything in the way of an interface spec to really start this discussion? Steve Richfield === On 7/3/08, William Pearson [EMAIL PROTECTED] wrote: Sorry about the long thread jack 2008/7/3 Vladimir Nesov [EMAIL PROTECTED]: On Thu, Jul 3, 2008 at 4:05 PM, William Pearson [EMAIL PROTECTED] wrote: Because it is dealing with powerful stuff, when it gets it wrong it goes wrong powerfully. You could lock the experimental code away in a sand box inside A, but then it would be a separate program just one inside A, but it might not be able to interact with programs in a way that it can do its job. There are two grades of faultiness. frequency and severity. You cannot predict the severity of faults of arbitrary programs (and accepting arbitrary programs from the outside world is something I want the system to be able to do, after vetting etc). You can't prove any interesting thing about an arbitrary program. It can behave like a Friendly AI before February 25, 2317, and like a Giant Cheesecake AI after that. Whoever said you could? The whole system is designed around the ability to take in or create arbitrary code, give it only minimal access to other programs that it can earn and lock it out from that ability when it does something bad. By arbitrary code I don't mean random, I mean stuff that has not formally been proven to have the properties you want. Formal proof is too high a burden to place on things that you want to win. You might not have the right axioms to prove the changes you want are right. Instead you can see the internals of the system as a form of continuous experiments. B is always testing a property of A or A', if at any time it stops having the property that B looks for then B flags it as buggy. I know this doesn't have the properties you would look for in a friendly AI set to dominate the world. But I think it is similar to the way humans work, and will be as chaotic and hard to grok as our neural structure. So as likely as humans are to explode intelligently. Will Pearson --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Will, Remember when I said that a purpose is not the same thing as a goal? The purpose that the system might be said to have embedded is attempting to maximise a certain signal. This purpose presupposes no ontology. The fact that this signal is attached to a human means the system as a whole might form the goal to try and please the human. Or depending on what the human does it might develop other goals. Goals are not the same as purposes. Goals require the intentional stance, purposes the design. To the extent that purpose is not related to goals, it is a meaningless term. In what possible sense is it worthwhile to talk about purpose if it doesn't somehow impact what an intelligent actually does? Possibly, but it will be another huge research topic to actually talk to the things that evolve in the artificial universe, as they will share very little background knowledge or ontology with us. I wish you luck and will be interested to see where you go but the alife route is just to slow and resource intensive for my liking. Will That is probably the most common criticism of the path I advocate and I certainly understand that, it's not for everyone. I will be very interested in your results as well, good luck! Terren --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Sorry about the late reply. snip some stuff sorted out 2008/6/30 Vladimir Nesov [EMAIL PROTECTED]: On Tue, Jul 1, 2008 at 2:02 AM, William Pearson [EMAIL PROTECTED] wrote: 2008/6/30 Vladimir Nesov [EMAIL PROTECTED]: If internals are programmed by humans, why do you need automatic system to assess them? It would be useful if you needed to construct and test some kind of combination/setting automatically, but not if you just test manually-programmed systems. How does the assessment platform help in improving/accelerating the research? Because to be interesting the human specified programs need to be autogenous, as in Josh Storr Hall's terminology, which means self-building. Capable of altering the stuff they are made of. In this case machine code equivalent. So you need the human to assess the improvements the system makes, for whatever purpose the human wants the system to perform. Altering the stuff they are made of is instrumental to achieving the goal, and should be performed where necessary, but it doesn't happen, for example, with individual brains. I think it happens at the level of neural structures. I.e. I think neural structures control the development of other neural structures. (I was planning to do the next blog post on this theme, maybe tomorrow.) Do you mean to create population of altered initial designs and somehow select from them (I hope not, it is orthogonal to what modification is for in the first place)? Otherwise, why do you still need automated testing? Could you present a more detailed use case? I'll try and give a fuller explanation later on. This means he needs to use a bunch more resources to get a singular useful system. Also the system might not do what he wants, but I don't think he minds about that. I'm allowing humans to design everything, just allowing the very low level to vary. Is this clearer? What do you mean by varying low level, especially in human-designed systems? The machine code the program is written in. Or in a java VM, the java bytecode. This still didn't make this point clearer. You can't vary the semantics of low-level elements from which software is built, and if you don't modify the semantics, any other modification is superficial and irrelevant. If it's not quite 'software' that you are running, and it is able to survive the modification of lower level, using the terms like 'machine code' and 'software' is misleading. And in any case, it's not clear what this modification of low level achieves. You can't extract work from obfuscation and tinkering, the optimization comes from the lawful and consistent pressure in the same direction. Okay let us clear things up. There are two things that need to be designed, a computer architecture or virtual machine and programs that form the initial set of programs within the system. Let us call the internal programs vmprograms to avoid confusion.The vmprograms should do all the heavy lifting (reasoning, creating new programs), this is where the lawlful and consistent pressure would come from. It is at source code of vmprograms that all needs to be changeable. However the pressure will have to be somewhat experimental to be powerful, you don't know what bugs a new program will have (if you are doing a non-tight proof search through the space of programs). So the point of the VM is to provide a safety net. If an experiment goes awry, then the VM should allow each program to limit the bugged vmprograms ability to affect it and eventually have it removed and the resources applied to it. Here is a toy scenario where the system needs this ability. *Note it is not anything that is like a full AI but illustrates a facet of something a full AI needs IMO*. Consider a system trying to solve a task, e.g. navigate a maze, that also has a number of different people out there giving helpful hints on how to solve the maze. These hints are in the form of patches to the vmprograms, e.g. changing the representation to 6-dimensional, giving another patch language that has better patches. So the system would make copies of the part of it to be patched and then patch it. Now you could give a patch evaluation module to see which patch works best, but what would happen if the module that implemented that vmprogram wanted to be patched? My solution to the problem is to allow the patch and non-patched version compete in the adhoc economic arena, and see which one wins. Does this clear things up? Will Pearson --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Terren, This is going too far. We can reconstruct to a considerable extent how humans think about problems - their conscious thoughts. Artists have been doing this reasonably well for hundreds of years. Science has so far avoided this, just as it avoided studying first the mind, with behaviourism, then consciousness,. The main reason cognitive science and psychology have avoided stream-of-thought studies (apart from v. odd scientists like Jerome Singer) is that conscious thought about problems is v. different from the highly ordered, rational, thinking of programmed computers which cog. sci. uses as its basic paradigm. In fact, human thinking is fundamentally different - the conscious self has major difficulty concentrating on any problem for any length of time - controlling the mind for more than a relatively few seconds, (as religious and humanistic thinkers have been telling us for thousands of years). Computers of course have perfect concentration forever. But that's because computers haven't had to deal with the type of problems that we do - the problematic problems where you don't, basically, know the answer, or how to find the answer, before you start. For this kind of problem - which is actually what differentiates AGI from narrow AI - human thinking, creative as opposed to rational, stumbling, scatty, and freely associative, is actually IDEAL, for all its imperfections. Yes, even if we extend our model of intelligence to include creative as well as rational thinking, it will still be an impoverished model, which may not include embodied thinking and perhaps other dimensions. But hey, we'll get there bit by bit, (just not, as we both agree, all at once in one five-year leap). Terren: My points about the pitfalls of theorizing about intelligence apply to any and all humans who would attempt it - meaning, it's not necessary to characterize AI folks in one way or another. There are any number of aspects of intelligence we could highlight that pose a challenge to orthodox models of intelligence, but the bigger point is that there are fundamental limits to the ability of an intelligence to observe itself, in exactly the same way that an eye cannot see itself. Consciousness and intelligence are present in every possible act of contemplation, so it is impossible to gain a vantage point of intelligence from outside of it. And that's exactly what we pretend to do when we conceptualize it within an artificial construct. This is the principle conceit of AI, that we can understand intelligence in an objective way, and model it well enough to reproduce by design. Terren --- On Tue, 7/1/08, Mike Tintner [EMAIL PROTECTED] wrote: Terren:It's to make the larger point that we may be so immersed in our own conceptualizations of intelligence - particularly because we live in our models and draw on our own experience and introspection to elaborate them - that we may have tunnel vision about the possibilities for better or different models. Or, we may take for granted huge swaths of what makes us so smart, because it's so familiar, or below the radar of our conscious awareness, that it doesn't even occur to us to reflect on it. No 2 is more relevant - AI-ers don't seem to introspect much. It's an irony that the way AI-ers think when creating a program bears v. little resemblance to the way programmed computers think. (Matt started to broach this when he talked a while back of computer programming as an art). But AI-ers seem to have no interest in the discrepancy - which again is ironic, because analysing it would surely help them with their programming as well as the small matter of understanding how general intelligence actually works. In fact - I just looked - there is a longstanding field on psychology of programming. But it seems to share the deficiency of psychology and cognitive science generally which is : no study of the stream-of-conscious-thought, especially conscious problemsolving. The only AI figure I know who did take some interest here was Herbert Simon who helped establish the use of verbal protocols. --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
On Wed, Jul 2, 2008 at 2:48 PM, William Pearson [EMAIL PROTECTED] wrote: Okay let us clear things up. There are two things that need to be designed, a computer architecture or virtual machine and programs that form the initial set of programs within the system. Let us call the internal programs vmprograms to avoid confusion.The vmprograms should do all the heavy lifting (reasoning, creating new programs), this is where the lawlful and consistent pressure would come from. It is at source code of vmprograms that all needs to be changeable. However the pressure will have to be somewhat experimental to be powerful, you don't know what bugs a new program will have (if you are doing a non-tight proof search through the space of programs). So the point of the VM is to provide a safety net. If an experiment goes awry, then the VM should allow each program to limit the bugged vmprograms ability to affect it and eventually have it removed and the resources applied to it. Here is a toy scenario where the system needs this ability. *Note it is not anything that is like a full AI but illustrates a facet of something a full AI needs IMO*. Consider a system trying to solve a task, e.g. navigate a maze, that also has a number of different people out there giving helpful hints on how to solve the maze. These hints are in the form of patches to the vmprograms, e.g. changing the representation to 6-dimensional, giving another patch language that has better patches. So the system would make copies of the part of it to be patched and then patch it. Now you could give a patch evaluation module to see which patch works best, but what would happen if the module that implemented that vmprogram wanted to be patched? My solution to the problem is to allow the patch and non-patched version compete in the adhoc economic arena, and see which one wins. What are the criteria that VM applies to vmprograms? If VM just shortcircuits the economic pressure of agents to one another, it in itself doesn't specify the direction of the search. The human economy works to efficiently satisfy the goals of human beings who already have their moral complexity. It propagates the decisions that customers make, and fuels the allocation of resources based on these decisions. Efficiency of economy is in efficiency of responding to information about human goals. If your VM just feeds the decisions on themselves, what stops the economy from focusing on efficiently doing nothing? -- Vladimir Nesov [EMAIL PROTECTED] http://causalityrelay.wordpress.com/ --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Mike, This is going too far. We can reconstruct to a considerable extent how humans think about problems - their conscious thoughts. Why is it going too far? I agree with you that we can reconstruct thinking, to a point. I notice you didn't say we can completely reconstruct how humans think about problems. Why not? We have two primary means for understanding thought, and both are deeply flawed: 1. Introspection. Introspection allows us to analyze our mental life in a reflective way. This is possible because we are able to construct mental models of our mental models. There are three flaws with introspection. The first, least serious flaw is that we only have access to that which is present in our conscious awareness. We cannot introspect about unconscious processes, by definition. This is a less serious objection because it's possible in practice to become conscious of phenomena there were previously unconscious, by developing our meta-mental-models. The question here becomes, is there any reason in principle that we cannot become conscious of *all* mental processes? The second flaw is that, because introspection relies on the meta-models we need to make sense of our internal, mental life, the possibility is always present that our meta-models themselves are flawed. Worse, we have no way of knowing if they are wrong, because we often unconsciously, unwittingly deny evidence contrary to our conception of our own cognition, particularly when it runs counter to a positive account of our self-image. Harvard's Project Implicit experiment (https://implicit.harvard.edu/implicit/) is a great way to demonstrate how we remain ignorant of deep, unconscious biases. Another example is how little we understand the contribution of emotion to our decision-making. Joseph Ledoux and others have shown fairly convincingly that emotion is a crucial part of human cognition, but most of us (particularly us men) deny the influence of emotion on our decision making. The final flaw is the most serious. It says there is a fundamental limit to what introspection has access to. This is the an eye cannot see itself objection. But I can see my eyes in the mirror, says the devil's advocate. Of course, a mirror lets us observe a reflected version of our eye, and this is what introspection is. But we cannot see inside our own eye, directly - it's a fundamental limitation of any observational apparatus. Likewise, we cannot see inside the very act of model-simulation that enables introspection. Introspection relies on meta-models, or models about models, which are activated/simulated *after the fact*. We might observe ourselves in the act of introspection, but that is nothing but a meta-meta-model. Each introspectional act by necessity is one step (at least) removed from the direct, in-the-present flow of cognition. This means that we can never observe the cognitive machinery that enables the act of introspection itself. And if you don't believe that introspection relies on cognitive machinery (maybe you're a dualist, but then why are you on an AI list? :-), ask yourself why we can't introspect about ourselves before a certain point in our young lives. It relies on a sufficiently sophisticated toolset that requires a certain amount of development before it is even possible. 2. Theory. Our theories of cognition are another path to understanding, and much of theory is directly or indirectly informed by introspection. When introspection fails (as in language acquisition), we rely completely on theory. The flaw with theory should be obvious. We have no direct way of testing theories of cognition, since we don't understand the connection between the mental and the physical. At best, we can use clever indirect means for generating evidence, and we usually have to accept the limits of reliability of subjective reports. Terren --- On Wed, 7/2/08, Mike Tintner [EMAIL PROTECTED] wrote: Terren, This is going too far. We can reconstruct to a considerable extent how humans think about problems - their conscious thoughts. Artists have been doing this reasonably well for hundreds of years. Science has so far avoided this, just as it avoided studying first the mind, with behaviourism, then consciousness,. The main reason cognitive science and psychology have avoided stream-of-thought studies (apart from v. odd scientists like Jerome Singer) is that conscious thought about problems is v. different from the highly ordered, rational, thinking of programmed computers which cog. sci. uses as its basic paradigm. In fact, human thinking is fundamentally different - the conscious self has major difficulty concentrating on any problem for any length of time - controlling the mind for more than a relatively few seconds, (as religious and humanistic thinkers have been telling us for thousands of years). Computers of course have perfect concentration forever.
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Terren, Obviously, as I indicated, I'm not suggesting that we can easily construct a total model of human cognition. But it ain't that hard to reconstruct reasonable and highly informative, if imperfect, models of how humans consciously think about problems. As I said, artists have been doing a reasonable job for centuries. Shakespeare, who really started the inner monologue, was arguably the first scientist of consciousness. The kind of standard argument you give below - the eye can't look at itself - is actually nonsense. Your conscious, inner thoughts are not that different from your public, recordable dialogue. (Any decent transcript of thought, BTW, will give a v. good indication of the emotions involved). We're not v. far apart here - we agree about the many dimensions of cognition, most of which are probably NOT directly accessible to the conscious mind. I'm just insisting on the massive importance of studying conscious thought. It was, as Crick said, ridiculous for science not to study consciousness - (it had a lot of rubbish arguments for not doing that, then) - it is equally ridiculous and in fact scientifically obscene not to study conscious thought. The consequences both for humans generally and AGI are enormous. Terren: Mike, This is going too far. We can reconstruct to a considerable extent how humans think about problems - their conscious thoughts. Why is it going too far? I agree with you that we can reconstruct thinking, to a point. I notice you didn't say we can completely reconstruct how humans think about problems. Why not? We have two primary means for understanding thought, and both are deeply flawed: 1. Introspection. Introspection allows us to analyze our mental life in a reflective way. This is possible because we are able to construct mental models of our mental models. There are three flaws with introspection. The first, least serious flaw is that we only have access to that which is present in our conscious awareness. We cannot introspect about unconscious processes, by definition. This is a less serious objection because it's possible in practice to become conscious of phenomena there were previously unconscious, by developing our meta-mental-models. The question here becomes, is there any reason in principle that we cannot become conscious of *all* mental processes? The second flaw is that, because introspection relies on the meta-models we need to make sense of our internal, mental life, the possibility is always present that our meta-models themselves are flawed. Worse, we have no way of knowing if they are wrong, because we often unconsciously, unwittingly deny evidence contrary to our conception of our own cognition, particularly when it runs counter to a positive account of our self-image. Harvard's Project Implicit experiment (https://implicit.harvard.edu/implicit/) is a great way to demonstrate how we remain ignorant of deep, unconscious biases. Another example is how little we understand the contribution of emotion to our decision-making. Joseph Ledoux and others have shown fairly convincingly that emotion is a crucial part of human cognition, but most of us (particularly us men) deny the influence of emotion on our decision making. The final flaw is the most serious. It says there is a fundamental limit to what introspection has access to. This is the an eye cannot see itself objection. But I can see my eyes in the mirror, says the devil's advocate. Of course, a mirror lets us observe a reflected version of our eye, and this is what introspection is. But we cannot see inside our own eye, directly - it's a fundamental limitation of any observational apparatus. Likewise, we cannot see inside the very act of model-simulation that enables introspection. Introspection relies on meta-models, or models about models, which are activated/simulated *after the fact*. We might observe ourselves in the act of introspection, but that is nothing but a meta-meta-model. Each introspectional act by necessity is one step (at least) removed from the direct, in-the-present flow of cognition. This means that we can never observe the cognitive machinery that enables the act of introspection itself. And if you don't believe that introspection relies on cognitive machinery (maybe you're a dualist, but then why are you on an AI list? :-), ask yourself why we can't introspect about ourselves before a certain point in our young lives. It relies on a sufficiently sophisticated toolset that requires a certain amount of development before it is even possible. 2. Theory. Our theories of cognition are another path to understanding, and much of theory is directly or indirectly informed by introspection. When introspection fails (as in language acquisition), we rely completely on theory. The flaw with theory should be obvious. We have no direct way of testing theories of cognition, since we don't understand the
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
2008/7/2 Terren Suydam [EMAIL PROTECTED]: Mike, This is going too far. We can reconstruct to a considerable extent how humans think about problems - their conscious thoughts. Why is it going too far? I agree with you that we can reconstruct thinking, to a point. I notice you didn't say we can completely reconstruct how humans think about problems. Why not? We have two primary means for understanding thought, and both are deeply flawed: 1. Introspection. Introspection allows us to analyze our mental life in a reflective way. This is possible because we are able to construct mental models of our mental models. There are three flaws with introspection. The first, least serious flaw is that we only have access to that which is present in our conscious awareness. We cannot introspect about unconscious processes, by definition. This is a less serious objection because it's possible in practice to become conscious of phenomena there were previously unconscious, by developing our meta-mental-models. The question here becomes, is there any reason in principle that we cannot become conscious of *all* mental processes? The second flaw is that, because introspection relies on the meta-models we need to make sense of our internal, mental life, the possibility is always present that our meta-models themselves are flawed. Worse, we have no way of knowing if they are wrong, because we often unconsciously, unwittingly deny evidence contrary to our conception of our own cognition, particularly when it runs counter to a positive account of our self-image. Harvard's Project Implicit experiment (https://implicit.harvard.edu/implicit/) is a great way to demonstrate how we remain ignorant of deep, unconscious biases. Another example is how little we understand the contribution of emotion to our decision-making. Joseph Ledoux and others have shown fairly convincingly that emotion is a crucial part of human cognition, but most of us (particularly us men) deny the influence of emotion on our decision making. The final flaw is the most serious. It says there is a fundamental limit to what introspection has access to. This is the an eye cannot see itself objection. But I can see my eyes in the mirror, says the devil's advocate. Of course, a mirror lets us observe a reflected version of our eye, and this is what introspection is. But we cannot see inside our own eye, directly - it's a fundamental limitation of any observational apparatus. Likewise, we cannot see inside the very act of model-simulation that enables introspection. Introspection relies on meta-models, or models about models, which are activated/simulated *after the fact*. We might observe ourselves in the act of introspection, but that is nothing but a meta-meta-model. Each introspectional act by necessity is one step (at least) removed from the direct, in-the-present flow of cognition. This means that we can never observe the cognitive machinery that enables the act of introspection itself. And if you don't believe that introspection relies on cognitive machinery (maybe you're a dualist, but then why are you on an AI list? :-), ask yourself why we can't introspect about ourselves before a certain point in our young lives. It relies on a sufficiently sophisticated toolset that requires a certain amount of development before it is even possible. 2. Theory. Our theories of cognition are another path to understanding, and much of theory is directly or indirectly informed by introspection. When introspection fails (as in language acquisition), we rely completely on theory. The flaw with theory should be obvious. We have no direct way of testing theories of cognition, since we don't understand the connection between the mental and the physical. At best, we can use clever indirect means for generating evidence, and we usually have to accept the limits of reliability of subjective reports. My plan is go for 3) Usefulness. Cognition is useful from an evolutionary point of view, if we try to create systems that are useful in the same situations (social, building world models), then we might one day stumble upon cognition. To expand on usefulness in social contexts, you have to ask yourself what the point of language is, why is it useful in an evolutionary setting. One thing the point of language is not, is fooling humans that you are human, which makes me annoyed at all the chatbots that get coverage as AI. I'll write more on this later. This by the way is why I don't self-organise purpose. I am pretty sure a specified purpose (not the same thing as a goal, at all) is needed for an intelligence. Will --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription:
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Mike, That's a rather weak reply. I'm open to the possibility that my ideas are incorrect or need improvement, but calling what I said nonsense without further justification is just hand waving. Unless you mean this as your justification: Your conscious, inner thoughts are not that different from your public, recordable dialogue. How this amounts to an objection to my points about introspection is beyond me... care to elaborate? Terren --- On Wed, 7/2/08, Mike Tintner [EMAIL PROTECTED] wrote: Terren, Obviously, as I indicated, I'm not suggesting that we can easily construct a total model of human cognition. But it ain't that hard to reconstruct reasonable and highly informative, if imperfect, models of how humans consciously think about problems. As I said, artists have been doing a reasonable job for centuries. Shakespeare, who really started the inner monologue, was arguably the first scientist of consciousness. The kind of standard argument you give below - the eye can't look at itself - is actually nonsense. Your conscious, inner thoughts are not that different from your public, recordable dialogue. (Any decent transcript of thought, BTW, will give a v. good indication of the emotions involved). We're not v. far apart here - we agree about the many dimensions of cognition, most of which are probably NOT directly accessible to the conscious mind. I'm just insisting on the massive importance of studying conscious thought. It was, as Crick said, ridiculous for science not to study consciousness - (it had a lot of rubbish arguments for not doing that, then) - it is equally ridiculous and in fact scientifically obscene not to study conscious thought. The consequences both for humans generally and AGI are enormous. Terren: Mike, This is going too far. We can reconstruct to a considerable extent how humans think about problems - their conscious thoughts. Why is it going too far? I agree with you that we can reconstruct thinking, to a point. I notice you didn't say we can completely reconstruct how humans think about problems. Why not? We have two primary means for understanding thought, and both are deeply flawed: 1. Introspection. Introspection allows us to analyze our mental life in a reflective way. This is possible because we are able to construct mental models of our mental models. There are three flaws with introspection. The first, least serious flaw is that we only have access to that which is present in our conscious awareness. We cannot introspect about unconscious processes, by definition. This is a less serious objection because it's possible in practice to become conscious of phenomena there were previously unconscious, by developing our meta-mental-models. The question here becomes, is there any reason in principle that we cannot become conscious of *all* mental processes? The second flaw is that, because introspection relies on the meta-models we need to make sense of our internal, mental life, the possibility is always present that our meta-models themselves are flawed. Worse, we have no way of knowing if they are wrong, because we often unconsciously, unwittingly deny evidence contrary to our conception of our own cognition, particularly when it runs counter to a positive account of our self-image. Harvard's Project Implicit experiment (https://implicit.harvard.edu/implicit/) is a great way to demonstrate how we remain ignorant of deep, unconscious biases. Another example is how little we understand the contribution of emotion to our decision-making. Joseph Ledoux and others have shown fairly convincingly that emotion is a crucial part of human cognition, but most of us (particularly us men) deny the influence of emotion on our decision making. The final flaw is the most serious. It says there is a fundamental limit to what introspection has access to. This is the an eye cannot see itself objection. But I can see my eyes in the mirror, says the devil's advocate. Of course, a mirror lets us observe a reflected version of our eye, and this is what introspection is. But we cannot see inside our own eye, directly - it's a fundamental limitation of any observational apparatus. Likewise, we cannot see inside the very act of model-simulation that enables introspection. Introspection relies on meta-models, or models about models, which are activated/simulated *after the fact*. We might observe ourselves in the act of introspection, but that is nothing but a meta-meta-model. Each introspectional act by necessity is one step (at least) removed from the direct, in-the-present flow of cognition. This means that we can never observe the cognitive machinery that enables the act of introspection itself. And if you don't believe that introspection
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Will, My plan is go for 3) Usefulness. Cognition is useful from an evolutionary point of view, if we try to create systems that are useful in the same situations (social, building world models), then we might one day stumble upon cognition. Sure, that's a valid approach for creating something we might call intelligent. My diatribe there was about human thought (the only kind we know of), not cognition in general. This by the way is why I don't self-organise purpose. I am pretty sure a specified purpose (not the same thing as a goal, at all) is needed for an intelligence. Will OK, then who or what specified the purpose of the first life forms? It's that intuition of yours that leads directly to Intelligent Design. As an aside, I love the irony that AI researchers who try to design intelligence are unwittingly giving ammunition to Intelligent Design arguments. Terren --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
2008/7/2 Vladimir Nesov [EMAIL PROTECTED]: On Wed, Jul 2, 2008 at 2:48 PM, William Pearson [EMAIL PROTECTED] wrote: Okay let us clear things up. There are two things that need to be designed, a computer architecture or virtual machine and programs that form the initial set of programs within the system. Let us call the internal programs vmprograms to avoid confusion.The vmprograms should do all the heavy lifting (reasoning, creating new programs), this is where the lawlful and consistent pressure would come from. It is at source code of vmprograms that all needs to be changeable. However the pressure will have to be somewhat experimental to be powerful, you don't know what bugs a new program will have (if you are doing a non-tight proof search through the space of programs). So the point of the VM is to provide a safety net. If an experiment goes awry, then the VM should allow each program to limit the bugged vmprograms ability to affect it and eventually have it removed and the resources applied to it. Here is a toy scenario where the system needs this ability. *Note it is not anything that is like a full AI but illustrates a facet of something a full AI needs IMO*. Consider a system trying to solve a task, e.g. navigate a maze, that also has a number of different people out there giving helpful hints on how to solve the maze. These hints are in the form of patches to the vmprograms, e.g. changing the representation to 6-dimensional, giving another patch language that has better patches. So the system would make copies of the part of it to be patched and then patch it. Now you could give a patch evaluation module to see which patch works best, but what would happen if the module that implemented that vmprogram wanted to be patched? My solution to the problem is to allow the patch and non-patched version compete in the adhoc economic arena, and see which one wins. What are the criteria that VM applies to vmprograms? If VM just shortcircuits the economic pressure of agents to one another, it in itself doesn't specify the direction of the search. The human economy works to efficiently satisfy the goals of human beings who already have their moral complexity. It propagates the decisions that customers make, and fuels the allocation of resources based on these decisions. Efficiency of economy is in efficiency of responding to information about human goals. If your VM just feeds the decisions on themselves, what stops the economy from focusing on efficiently doing nothing? They would get less credit from the human supervisor. Let me expand on what I meant about the economic competition. Let us say vmprogram A makes a copy of itself, called A', with some purposeful tweaks, trying to make itself more efficient. A' has some bugs such that the human notices something wrong with the system, she gives less credit on average each time A' is helping out rather than A. Now A and A' both have to bid for the chance to help program B which is closer to the outputting (due to the programming of B), B pays a proportion of the credit it gets back. Now the credit B gets will be lower when A' is helping, than when A is helping. So A' will get less in general than A. There are a few scenarios, ordered from quickest acting to slowest. 1 ) B keeps records of who helps him and sees that A' is not helping him as well as the average, so no longer lets A' bid. A' resources get used when it can't keep up bidding for them. 2) A' continues bidding a lot, to outbid A. However the average amount A' gets is less than it gets back from B. A' bankrupts itself and other programs use its resources. 3) A' doesn't manage to outbid A' after a fair few trials, so gets the same fate as it does in scenario 1) If you start with a bunch of stupid vmprograms, you won't get anywhere. It can just go to nothingness, you do have to design them fairly well, just in such a way that that design can change later. Will --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
How do you assign credit to programs that are good at generating good children? Particularly, could a program specialize in this, so that it doesn't do anything useful directly but always through making highly useful children? On Wed, Jul 2, 2008 at 1:09 PM, William Pearson [EMAIL PROTECTED] wrote: 2008/7/2 Vladimir Nesov [EMAIL PROTECTED]: On Wed, Jul 2, 2008 at 2:48 PM, William Pearson [EMAIL PROTECTED] wrote: Okay let us clear things up. There are two things that need to be designed, a computer architecture or virtual machine and programs that form the initial set of programs within the system. Let us call the internal programs vmprograms to avoid confusion.The vmprograms should do all the heavy lifting (reasoning, creating new programs), this is where the lawlful and consistent pressure would come from. It is at source code of vmprograms that all needs to be changeable. However the pressure will have to be somewhat experimental to be powerful, you don't know what bugs a new program will have (if you are doing a non-tight proof search through the space of programs). So the point of the VM is to provide a safety net. If an experiment goes awry, then the VM should allow each program to limit the bugged vmprograms ability to affect it and eventually have it removed and the resources applied to it. Here is a toy scenario where the system needs this ability. *Note it is not anything that is like a full AI but illustrates a facet of something a full AI needs IMO*. Consider a system trying to solve a task, e.g. navigate a maze, that also has a number of different people out there giving helpful hints on how to solve the maze. These hints are in the form of patches to the vmprograms, e.g. changing the representation to 6-dimensional, giving another patch language that has better patches. So the system would make copies of the part of it to be patched and then patch it. Now you could give a patch evaluation module to see which patch works best, but what would happen if the module that implemented that vmprogram wanted to be patched? My solution to the problem is to allow the patch and non-patched version compete in the adhoc economic arena, and see which one wins. What are the criteria that VM applies to vmprograms? If VM just shortcircuits the economic pressure of agents to one another, it in itself doesn't specify the direction of the search. The human economy works to efficiently satisfy the goals of human beings who already have their moral complexity. It propagates the decisions that customers make, and fuels the allocation of resources based on these decisions. Efficiency of economy is in efficiency of responding to information about human goals. If your VM just feeds the decisions on themselves, what stops the economy from focusing on efficiently doing nothing? They would get less credit from the human supervisor. Let me expand on what I meant about the economic competition. Let us say vmprogram A makes a copy of itself, called A', with some purposeful tweaks, trying to make itself more efficient. A' has some bugs such that the human notices something wrong with the system, she gives less credit on average each time A' is helping out rather than A. Now A and A' both have to bid for the chance to help program B which is closer to the outputting (due to the programming of B), B pays a proportion of the credit it gets back. Now the credit B gets will be lower when A' is helping, than when A is helping. So A' will get less in general than A. There are a few scenarios, ordered from quickest acting to slowest. 1 ) B keeps records of who helps him and sees that A' is not helping him as well as the average, so no longer lets A' bid. A' resources get used when it can't keep up bidding for them. 2) A' continues bidding a lot, to outbid A. However the average amount A' gets is less than it gets back from B. A' bankrupts itself and other programs use its resources. 3) A' doesn't manage to outbid A' after a fair few trials, so gets the same fate as it does in scenario 1) If you start with a bunch of stupid vmprograms, you won't get anywhere. It can just go to nothingness, you do have to design them fairly well, just in such a way that that design can change later. Will --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
2008/7/2 Abram Demski [EMAIL PROTECTED]: How do you assign credit to programs that are good at generating good children? I never directly assign credit, apart from the first stage. The rest of the credit assignment is handled by the vmprograms, er, programming. Particularly, could a program specialize in this, so that it doesn't do anything useful directly but always through making highly useful children? As the parent controls the code of its offspring, it could embed code in its offspring to pass a small portion of the credit they get back to it. They would have to be careful how much to skim off so the offspring could still thrive. Will --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
2008/7/2 Vladimir Nesov [EMAIL PROTECTED]: On Wed, Jul 2, 2008 at 9:09 PM, William Pearson [EMAIL PROTECTED] wrote: They would get less credit from the human supervisor. Let me expand on what I meant about the economic competition. Let us say vmprogram A makes a copy of itself, called A', with some purposeful tweaks, trying to make itself more efficient. So, this process performs optimization, A has a goal that it tries to express in form of A'. What is the problem with the algorithm that A uses? If this algorithm is stupid (in a technical sense), A' is worse than A and we can detect that. But this means that in fact, A' doesn't do its job and all the search pressure comes from program B that ranks the performance of A or A'. This generate-blindly-or-even-stupidly-and-check is a very inefficient algorithm. If, on the other hand, A happens to be a good program, then A' has a good change of being better than A, and anyway A has some understanding of what 'better' means, then what is the role of B? B adds almost no additional pressure, almost everything is done by A. How do you distribute the optimization pressure between generating programs (A) and checking programs (B)? Why do you need to do that at all, what is the benefit of generating and checking separately, compared to reliably generating from the same point (A alone)? If generation is not reliable enough, it probably won't be useful as optimization pressure anyway. The point of A and A' is that A', if better, may one day completely replace A. What is very good? Is 1 in 100 chances of making a mistake when generating its successor very good? If you want A' to be able to replace A, that is only 100 generations before you have made a bad mistake, and then where do you go? You have a bugged program and nothing to act as a watchdog. Also if A' is better than time A at time t, there is no guarantee that it will stay that way. Changes in the environment might favour one optimisation over another. If they both do things well, but different things then both A and A' might survive in different niches. I would also be interested in why you think we have programmers and system testers in the real world. Also worth noting is most optimisation will be done inside the vmprograms, this process is only for very fundamental code changes, e.g. changing representations, biases, ways of creating offspring. Things that cannot be tested easily any other way. I'm quite happy for it to be slow, because this process is not where the majority of quickness of the system will rest. But this process is needed for intelligence else you will be stuck with certain ways of doing things when they are not useful. Will Pearson --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
On Thu, Jul 3, 2008 at 12:59 AM, William Pearson [EMAIL PROTECTED] wrote: 2008/7/2 Vladimir Nesov [EMAIL PROTECTED]: On Wed, Jul 2, 2008 at 9:09 PM, William Pearson [EMAIL PROTECTED] wrote: They would get less credit from the human supervisor. Let me expand on what I meant about the economic competition. Let us say vmprogram A makes a copy of itself, called A', with some purposeful tweaks, trying to make itself more efficient. So, this process performs optimization, A has a goal that it tries to express in form of A'. What is the problem with the algorithm that A uses? If this algorithm is stupid (in a technical sense), A' is worse than A and we can detect that. But this means that in fact, A' doesn't do its job and all the search pressure comes from program B that ranks the performance of A or A'. This generate-blindly-or-even-stupidly-and-check is a very inefficient algorithm. If, on the other hand, A happens to be a good program, then A' has a good change of being better than A, and anyway A has some understanding of what 'better' means, then what is the role of B? B adds almost no additional pressure, almost everything is done by A. How do you distribute the optimization pressure between generating programs (A) and checking programs (B)? Why do you need to do that at all, what is the benefit of generating and checking separately, compared to reliably generating from the same point (A alone)? If generation is not reliable enough, it probably won't be useful as optimization pressure anyway. The point of A and A' is that A', if better, may one day completely replace A. What is very good? Is 1 in 100 chances of making a mistake when generating its successor very good? If you want A' to be able to replace A, that is only 100 generations before you have made a bad mistake, and then where do you go? You have a bugged program and nothing to act as a watchdog. Also if A' is better than time A at time t, there is no guarantee that it will stay that way. Changes in the environment might favour one optimisation over another. If they both do things well, but different things then both A and A' might survive in different niches. I suggest you read ( http://sl4.org/wiki/KnowabilityOfFAI ) If your program is a faulty optimizer that can't pump the reliability out of its optimization, you are doomed. I assume you argue that you don't want to include B in A, because a descendant of A may start to fail unexpectedly. But if you reliably copy B inside each of A's descendants, this particular problem won't appear. The main question is: what is the difference between just trying to build a self-improving program A and doing so inside your testing environment. If there is no difference, you add nothing by your framework. If there is, it would be good to find out what it is. I would also be interested in why you think we have programmers and system testers in the real world. Testing that doesn't even depend on program's internal structure and only checks its output (as in your economy setup) isn't nearly good enough. Testing that you're referring to in this post (activity performed by humans, based on specific implementation and understanding of high-level specification that says what algorithm should do) has very little to do with testing that you propose in the framework (fixed program B). Anyway, you should answer on that question yourself: what is the essence of useful activity that is performed by software testing and that you capture in your framework. Arguing that there must be some such essence and that it must transfer to your setting isn't reliable. Also worth noting is most optimisation will be done inside the vmprograms, this process is only for very fundamental code changes, e.g. changing representations, biases, ways of creating offspring. Things that cannot be tested easily any other way. I'm quite happy for it to be slow, because this process is not where the majority of quickness of the system will rest. But this process is needed for intelligence else you will be stuck with certain ways of doing things when they are not useful. Being stuck in development is a problem of search process, it can as well be a problem of process A that should be resolved from within A. -- Vladimir Nesov [EMAIL PROTECTED] http://causalityrelay.wordpress.com/ --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: Savants and user-interfaces [was Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
2008/6/30 Vladimir Nesov [EMAIL PROTECTED]: On Tue, Jul 1, 2008 at 8:31 AM, Linas Vepstas [EMAIL PROTECTED] wrote: Why binary? I once skimmed a biography of Ramanujan, he started multiplying numbers in his head as a pre-teen. I suspect it was grindingly boring, but given the surroundings, might have been the most fun thing he could think of. If you're autistic, then focusing obsessively on some task might be a great way to pass the time, but if you're more or less normal, I doubt you'll get very far with obsessive-compulsive self-training -- and that's the problem, isn't it? If the signals have properties of their own, I'm afraid they will start interfering with each other, which won't allow the circuit to execute in real time. Binary signals, on the other hand, can be encoded by the activation of nodes of the circuit, active/inactive. If you have an AND gate that leads from symbols S1 and S2 to S3, you learn to remember S3 only when you see both S1 and S2 What are you trying to accomplish here? I don't see where you are trying to go with this. I don't think a human can consciously train one or two neurons to do something, we train millions at a time. -- I'm guessing savants only employ a few tens of million neurons (give or take a few orders of magnitude) -- to do their stuff. Still, an array of 1K by 1K electrodes is well within current technology, we just don't know where to hook this up to, with the exception of simple motor areas, retina, and bit of the auditory circuits. --linas --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: Savants and user-interfaces [was Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
On Tue, Jul 1, 2008 at 10:02 AM, Linas Vepstas [EMAIL PROTECTED] wrote: What are you trying to accomplish here? I don't see where you are trying to go with this. I don't think a human can consciously train one or two neurons to do something, we train millions at a time. -- I'm guessing savants only employ a few tens of million neurons (give or take a few orders of magnitude) -- to do their stuff. Still, an array of 1K by 1K electrodes is well within current technology, we just don't know where to hook this up to, with the exception of simple motor areas, retina, and bit of the auditory circuits. Certainly nothing to do with individual neurons. Basically, it's possible to train a finite state automaton in the mind through association. You see a certain combination of properties, you think the symbol that describes this combination. If such automaton is trained not just to handle natural data (such as language), but to a specifically designed circuit plan, it'll probably be possible to use it as a directly accessible 'add-on' to the brain that implements specific simple function efficiently, such as some operation with numbers using a clever algorithm in a way alien to normal deliberative learning. You don't learn to perform a task, but to execute individual steps of an algorithm that performs a task. -- Vladimir Nesov [EMAIL PROTECTED] http://causalityrelay.wordpress.com/ --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: Savants and user-interfaces [was Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
I was nearly kicked out of school in seventh grade for coming up with a method of manipulating (multiplying, dividing) large numbers in my head using what I later learned was a shift-reduce method. It was similar to this: http://www.metacafe.com/watch/742717/human_calculator/ My seventh grade math teacher was so upset with me, he almost struck me (physically -- you could get away with that back them). His reason? Wasting valuable math class time. The point is, you can train yourself to do this type of thing and look very savant-like. The above link is just one in a series of videos where the teacher presents this system. It takes practice, but not much more than learning the standard multiplication table. Cheers, Brad Vladimir Nesov wrote: Interesting: is it possible to train yourself to run a specially designed nontrivial inference circuit based on low-base transformations (e.g. binary)? You start by assigning unique symbols to its nodes, train yourself to stably perform associations implementing its junctions, and then assemble it all together by training yourself to generate a problem as a temporal sequence (request), so that it can be handled by the overall circuit, and training to read out the answer and convert it to sequence of e.g. base-10 digits or base-100 words keying pairs of digits (like in mnemonic)? Has anyone heard of this attempted? At least the initial steps look straightforward enough, what kind of obstacles this kind of experiment can run into? On Tue, Jul 1, 2008 at 7:43 AM, Linas Vepstas [EMAIL PROTECTED] wrote: 2008/6/30 Terren Suydam [EMAIL PROTECTED]: savant I've always theorized that savants can do what they do because they've been able to get direct access to, and train, a fairly small number of neurons in their brain, to accomplish highly specialized (and thus rather unusual) calculations. I'm thinking specifically of Ramanujan, the Hindi mathematician. He appears to have had access to a multiply-add type circuit in his brain, and could do symbolic long division and multiplication as a result -- I base this on studying some of the things he came up with -- after a while, it seems to be clear how he came up with it (even if the feat is clearly not reproducible). In a sense, similar feats are possible by using a modern computer with a good algebra system. Simon Plouffe seems to be a modern-day example of this: he noodles around with his systems, and finds various interesting relationships that would otherwise be obscure/unknown. He does this without any particularly deep or expansive training in math (whence some of his friction with real academics). If Simon could get a computer-algebra chip implanted in his brain, (i.e. with a very, very user-freindly user-interface) so that he could work the algebra system just by thinking about it, I bet his output would resemble that of Ramanujan a whole lot more than it already does -- as it were, he's hobbled by a crappy user interface. Thus, let me theorize: by studying savants with MRI and what-not, we may find a way of getting a much better man-machine interface. That is, currently, electrodes are always implanted in motor neurons (or visual cortex, etc) i.e. in places of the brain with very low levels of abstraction from the real word. It would be interesting to move up the level of abstraction, and I think that studying how savants access the magic circuits in thier brain will open up a method for high-level interfaces to external computing machinery. --linas --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
2008/6/30 Terren Suydam [EMAIL PROTECTED]: Hi Will, --- On Mon, 6/30/08, William Pearson [EMAIL PROTECTED] wrote: The only way to talk coherently about purpose within the computation is to simulate self-organized, embodied systems. I don't think you are quite getting my system. If you had a bunch of programs that did the following 1) created new programs, by trial and error and taking statistics of variables or getting arbitrary code from the outside. 2) communicated with each other to try and find programs that perform services they need. 3) Bid for computer resources, if a program loses its memory resources it is selected against, in a way. Would this be sufficiently self-organised? If not, why not? And the computer programs would be as embodied as your virtual creatures. They would just be embodied within a tacit economy, rather than an artificial chemistry. It boils down to your answer to the question: how are the resources ultimately allocated to the programs? If you're the one specifying it, via some heuristic or rule, then the purpose is driven by you. If resource allocation is handled by some self-organizing method (this wasn't clear in the article you provided), then I'd say that the system's purpose is self-defined. I'm not sure how the system qualifies. It seems to be half way between the two definitions you gave. The programs can have special instructions in that bid for a specific resource with as much credit as they want (see my recent message replying to Vladimir Nesov for more information about banks, bidding and credit). The instructions can be removed or not done, the amount of credit bid can be changed. The credit is given to some programs by a fixed function, but they have instructions they can execute (or not) to give it to other programs forming an economy. What say you, self-organised or not? As for embodiment, my question is, how do your programs receive input? Embodiment, as I define it, requires that inputs are merely reflections of state variables, and not even labeled in any way... i.e. we can't pre-define ontologies. The embodied entity starts from the most unstructured state possible and self-structures whatever inputs it receives. Bits and bytes from the outside world, or bits and bytes from reading other programs programing and data. No particular ontology. That said, you may very well be doing that and be creating embodied programs in this way... if so, that's cool because I hadn't considered that possibility and I'll be interested to see how you fare. It is going to take a while. Virtual machine writing is very unrewarding programming. I have other things to do right now, I'll get back to the rest of the message in a bit. Will Pearson --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Terren:It's to make the larger point that we may be so immersed in our own conceptualizations of intelligence - particularly because we live in our models and draw on our own experience and introspection to elaborate them - that we may have tunnel vision about the possibilities for better or different models. Or, we may take for granted huge swaths of what makes us so smart, because it's so familiar, or below the radar of our conscious awareness, that it doesn't even occur to us to reflect on it. No 2 is more relevant - AI-ers don't seem to introspect much. It's an irony that the way AI-ers think when creating a program bears v. little resemblance to the way programmed computers think. (Matt started to broach this when he talked a while back of computer programming as an art). But AI-ers seem to have no interest in the discrepancy - which again is ironic, because analysing it would surely help them with their programming as well as the small matter of understanding how general intelligence actually works. In fact - I just looked - there is a longstanding field on psychology of programming. But it seems to share the deficiency of psychology and cognitive science generally which is : no study of the stream-of-conscious-thought, especially conscious problemsolving. The only AI figure I know who did take some interest here was Herbert Simon who helped establish the use of verbal protocols. --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: Savants and user-interfaces [was Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
2008/7/1 Vladimir Nesov [EMAIL PROTECTED]: On Tue, Jul 1, 2008 at 10:02 AM, Linas Vepstas [EMAIL PROTECTED] wrote: What are you trying to accomplish here? I don't see where you are trying to go with this. I don't think a human can consciously train one or two neurons to do something, we train millions at a time. -- I'm guessing savants only employ a few tens of million neurons (give or take a few orders of magnitude) -- to do their stuff. Still, an array of 1K by 1K electrodes is well within current technology, we just don't know where to hook this up to, with the exception of simple motor areas, retina, and bit of the auditory circuits. Certainly nothing to do with individual neurons. Basically, it's possible to train a finite state automaton in the mind through association. You see a certain combination of properties, you think the symbol that describes this combination. If such automaton is trained not just to handle natural data (such as language), but to a specifically designed circuit plan, it'll probably be possible to use it as a directly accessible 'add-on' to the brain that implements specific simple function efficiently, such as some operation with numbers using a clever algorithm in a way alien to normal deliberative learning. You don't learn to perform a task, but to execute individual steps of an algorithm that performs a task. Yes, but isn't the interesting case in the other direction? We have ordinary computers that can already do quite well computationally. What we *don't* have a a good man-machine interface. For example, modern disk drives hold more bytes than the human mind can. I don't want to train myself for feats of memorization, I want automatic and instant access to a disk drive. So, perhaps by studying savants who are capable of memorization feats, perhaps we can find the sort of neural circuitry needed to interface to a disk drive. It is, perhaps because savants have these unusual abilities, that it sheds light on the kind of wiring that would be needed for electrodes. --linas --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Will, I think the original issue was about purpose. In your system, since a human is the one determining which programs are performing the best, the purpose is defined in the mind of the human. Beyond that, it certainly sounds as if it is a self-organizing system. Terren --- On Tue, 7/1/08, William Pearson [EMAIL PROTECTED] wrote: I'm not sure how the system qualifies. It seems to be half way between the two definitions you gave. The programs can have special instructions in that bid for a specific resource with as much credit as they want (see my recent message replying to Vladimir Nesov for more information about banks, bidding and credit). The instructions can be removed or not done, the amount of credit bid can be changed. The credit is given to some programs by a fixed function, but they have instructions they can execute (or not) to give it to other programs forming an economy. What say you, self-organised or not? --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Hi Mike, My points about the pitfalls of theorizing about intelligence apply to any and all humans who would attempt it - meaning, it's not necessary to characterize AI folks in one way or another. There are any number of aspects of intelligence we could highlight that pose a challenge to orthodox models of intelligence, but the bigger point is that there are fundamental limits to the ability of an intelligence to observe itself, in exactly the same way that an eye cannot see itself. Consciousness and intelligence are present in every possible act of contemplation, so it is impossible to gain a vantage point of intelligence from outside of it. And that's exactly what we pretend to do when we conceptualize it within an artificial construct. This is the principle conceit of AI, that we can understand intelligence in an objective way, and model it well enough to reproduce by design. Terren --- On Tue, 7/1/08, Mike Tintner [EMAIL PROTECTED] wrote: Terren:It's to make the larger point that we may be so immersed in our own conceptualizations of intelligence - particularly because we live in our models and draw on our own experience and introspection to elaborate them - that we may have tunnel vision about the possibilities for better or different models. Or, we may take for granted huge swaths of what makes us so smart, because it's so familiar, or below the radar of our conscious awareness, that it doesn't even occur to us to reflect on it. No 2 is more relevant - AI-ers don't seem to introspect much. It's an irony that the way AI-ers think when creating a program bears v. little resemblance to the way programmed computers think. (Matt started to broach this when he talked a while back of computer programming as an art). But AI-ers seem to have no interest in the discrepancy - which again is ironic, because analysing it would surely help them with their programming as well as the small matter of understanding how general intelligence actually works. In fact - I just looked - there is a longstanding field on psychology of programming. But it seems to share the deficiency of psychology and cognitive science generally which is : no study of the stream-of-conscious-thought, especially conscious problemsolving. The only AI figure I know who did take some interest here was Herbert Simon who helped establish the use of verbal protocols. --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Hi Ben, I don't think the flaw you have identified matters to the main thrust of Richard's argument - and if you haven't summarized Richard's position precisely, you have summarized mine. :-] You're saying the flaw in that position is that prediction of complex networks might merely be a matter of computational difficulty, rather than fundamentally intractability. But any formally defined complex system is going to be computable in principle. We can always predict such a system with infinite computing power. That doesn't make it tractable, or open to understanding, because obviously real understanding can't be dependent infinite computing power. The question of fundamental intractability comes down to the degree with which we can make predictions about the global level from the local. And let's hope there's progress to be made there because each discovery will make our lives easier, to those of us who would try to understand something like the brain or the body or even just the cell. Or even just folding proteins! But it seems pretty obvious to me anyway that we will never be able to predict the weather with any precision without doing an awful lot of computation. And what is our mind but the weather in our brains? Terren --- On Sun, 6/29/08, Ben Goertzel [EMAIL PROTECTED] wrote: From: Ben Goertzel [EMAIL PROTECTED] Subject: Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI To: agi@v2.listbox.com Date: Sunday, June 29, 2008, 10:44 PM Richard, I think that it would be possible to formalize your complex systems argument mathematically, but I don't have time to do so right now. Or, then again . perhaps I am wrong: maybe you really *cannot* understand anything except math? It's not the case that I can only understand math -- however, I have a lot of respect for the power of math to clarify disagreements. Without math, arguments often proceed in a confused way because different people are defining terms differently,a and don't realize it. But, I agree math is not the only kind of rigor. I would be happy with a very careful, systematic exposition of your argument along the lines of Spinoza or the early Wittgenstein. Their arguments were not mathematical, but were very rigorous and precisely drawn -- not slippery. Perhaps you have no idea what the actual argument is, and that has been the problem all along? I notice that you avoided answering my request that you summarize your argument against the complex systems problem ... perhaps you are just confused about what the argument actually is, and have been confused right from the beginning? In a nutshell, it seems you are arguing that general intelligence is fundamentally founded on emergent properties of complex systems, and that it's not possible for us to figure out analytically how these emergent properties emerge from the lower-level structures and dynamics of the complex systems involved. Evolution, you suggest, figured out some complex systems that give rise to the appropriate emergent properties to produce general intelligence. But evolution did not do this figuring-out in an analytical way, rather via its own special sort of directed trial and error. You suggest that to create a generally intelligent system, we should create a software framework that makes it very easy to experiment with different sorts of complex systems, so that we can then figure out (via some combination of experiment, analysis, intuition, theory, etc.) how to create a complex system that gives rise to the emergent properties associated with general intelligence. I'm sure the above is not exactly how you'd phrase your argument -- and it doesn't capture all the nuances -- but I was trying to give a compact and approximate formulation. If you'd like to give an alternative, equally compact formulation, that would be great. I think the flaw of your argument lies in your definition of complexity, and that this would be revealed if you formalized your argument more fully. I think you define complexity as a kind of fundamental irreducibility that the human brain does not possess, and that engineered AGI systems need not possess. I think that real systems display complexity which makes it **computationally difficult** to explain their emergent properties in terms of their lower-level structures and dynamics, but not as fundamentally intractable as you presume. But because you don't formalize your notion of complexity adequately, it's not possible to engage you in rational argumentation regarding the deep flaw at the center of your argument. However, I cannot prove rigorously that the brain is NOT complex in the overly strong sense you allude it is ... and nor can I prove rigorously that a design like Novamente Cognition Engine or OpenCog Prime will give rise to the emergent properties
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
But, we don't need to be able to predict the thoughts of an AGI system in detail, to be able to architect an AGI system that has thoughts... I agree that predicting the thoughts of an AGI system in detail is going to be pragmatically impossible ... but I don't agree that predicting **which** AGI designs can lead to the emergent properties corresponding to general intelligence, is pragmatically impossible to do in an analytical and rational way ... Similarly, I could engineer an artificial weather system displaying hurricanes, whirlpools, or whatever phenomena you ask me for -- based on my general understanding of the Navier-stokes equation. Even though I could not, then, predict the specific dynamics of those hurricanes, whirlpools, etc. We lack the equivalent of the Navier-stokes equation for thoughts. But we can still arrive at reasonable analytic understandings of appropriately constrained and formalised AGI designs, with the power to achieve general intelligence... ben g On Mon, Jun 30, 2008 at 1:55 AM, Terren Suydam [EMAIL PROTECTED] wrote: Hi Ben, I don't think the flaw you have identified matters to the main thrust of Richard's argument - and if you haven't summarized Richard's position precisely, you have summarized mine. :-] You're saying the flaw in that position is that prediction of complex networks might merely be a matter of computational difficulty, rather than fundamentally intractability. But any formally defined complex system is going to be computable in principle. We can always predict such a system with infinite computing power. That doesn't make it tractable, or open to understanding, because obviously real understanding can't be dependent infinite computing power. The question of fundamental intractability comes down to the degree with which we can make predictions about the global level from the local. And let's hope there's progress to be made there because each discovery will make our lives easier, to those of us who would try to understand something like the brain or the body or even just the cell. Or even just folding proteins! But it seems pretty obvious to me anyway that we will never be able to predict the weather with any precision without doing an awful lot of computation. And what is our mind but the weather in our brains? Terren --- On Sun, 6/29/08, Ben Goertzel [EMAIL PROTECTED] wrote: From: Ben Goertzel [EMAIL PROTECTED] Subject: Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI To: agi@v2.listbox.com Date: Sunday, June 29, 2008, 10:44 PM Richard, I think that it would be possible to formalize your complex systems argument mathematically, but I don't have time to do so right now. Or, then again . perhaps I am wrong: maybe you really *cannot* understand anything except math? It's not the case that I can only understand math -- however, I have a lot of respect for the power of math to clarify disagreements. Without math, arguments often proceed in a confused way because different people are defining terms differently,a and don't realize it. But, I agree math is not the only kind of rigor. I would be happy with a very careful, systematic exposition of your argument along the lines of Spinoza or the early Wittgenstein. Their arguments were not mathematical, but were very rigorous and precisely drawn -- not slippery. Perhaps you have no idea what the actual argument is, and that has been the problem all along? I notice that you avoided answering my request that you summarize your argument against the complex systems problem ... perhaps you are just confused about what the argument actually is, and have been confused right from the beginning? In a nutshell, it seems you are arguing that general intelligence is fundamentally founded on emergent properties of complex systems, and that it's not possible for us to figure out analytically how these emergent properties emerge from the lower-level structures and dynamics of the complex systems involved. Evolution, you suggest, figured out some complex systems that give rise to the appropriate emergent properties to produce general intelligence. But evolution did not do this figuring-out in an analytical way, rather via its own special sort of directed trial and error. You suggest that to create a generally intelligent system, we should create a software framework that makes it very easy to experiment with different sorts of complex systems, so that we can then figure out (via some combination of experiment, analysis, intuition, theory, etc.) how to create a complex system that gives rise to the emergent properties associated with general intelligence. I'm sure the above is not exactly how you'd phrase your argument -- and it doesn't capture all the nuances -- but I was trying to give a compact and approximate formulation. If you'd like to give
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
On Mon, Jun 30, 2008 at 8:07 AM, Terren Suydam [EMAIL PROTECTED] wrote: By the way, just wanted to point out a beautifully simple example - perhaps the simplest - of an irreducibility in complex systems. Individual molecular interactions are symmetric in time, they work the same forwards and backwards. Yet diffusion, which is nothing more than the aggregate of molecular interactions, is asymmetric. Figure that one out. This is just statistical mechanics. The interesting thing is that we make an opportunistic assumption, that any colliding particles are independent before collision (this introduces the time arrow), which is then empirically confirmed by the fact that derived properties agree with the phenomenological theory of entropy. --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
--- On Mon, 6/30/08, Ben Goertzel [EMAIL PROTECTED] wrote: but I don't agree that predicting **which** AGI designs can lead to the emergent properties corresponding to general intelligence, is pragmatically impossible to do in an analytical and rational way ... OK, I grant you that you may be able to do that. I believe that we can be extremely clever in this regard. An example of that is an implementation of a Turing Machine within the Game of Life: http://rendell-attic.org/gol/tm.htm What a beautiful construction. But it's completely contrived. What you're suggesting is equivalent, because your design is contrived by your own intelligence. [I understand that within the Novamente idea is room for non-deterministic (for practical purposes) behavior, so it doesn't suffer from the usual complexity-inspired criticisms of purely logical systems.] But whatever achievement you make, it's just one particular design that may prove effective in some set of domains. And there's the rub - the fact that your design is at least partially static will limit its applicability in some set of domains. I make this argument more completely here: http://www.machineslikeus.com/cms/news/design-bad-or-why-artificial-intelligence-needs-artificial-life or http://tinyurl.com/3coavb If you design a robot, you limit its degrees of freedom. And there will be environments it cannot get around in. By contrast, if you have a design that is capable of changing itself (even if that means from generation to generation), then creative configurations can be discovered. The same basic idea works in the mental arena as well. If you specify the mental machinery, there will be environments it cannot get around in, so to speak. There will be important ways in which it is unable to adapt. You are limiting your design by your own intelligence, which though considerable, is no match for the creativity manifest in a single biological cell. Terren --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
I agree that all designed systems have limitations, but I also suggest that all evolved systems have limitations. This is just the no free lunch theorem -- in order to perform better than random search at certain optimization tasks, a system needs to have some biases built in, and these biases will cause it to work WORSE than random search on some other optimization tasks. No AGI based on finite resources will ever be **truly** general, be it an engineered or evolved systems Evolved systems are far from being beyond running into dead ends ... their adaptability is far from infinite ... the evolutionary process itself may be endlessly creative, but in that sense so may be the self-modifying process of an engineered AGI ... -- Ben G On Mon, Jun 30, 2008 at 3:17 AM, Terren Suydam [EMAIL PROTECTED] wrote: --- On Mon, 6/30/08, Ben Goertzel [EMAIL PROTECTED] wrote: but I don't agree that predicting **which** AGI designs can lead to the emergent properties corresponding to general intelligence, is pragmatically impossible to do in an analytical and rational way ... OK, I grant you that you may be able to do that. I believe that we can be extremely clever in this regard. An example of that is an implementation of a Turing Machine within the Game of Life: http://rendell-attic.org/gol/tm.htm What a beautiful construction. But it's completely contrived. What you're suggesting is equivalent, because your design is contrived by your own intelligence. [I understand that within the Novamente idea is room for non-deterministic (for practical purposes) behavior, so it doesn't suffer from the usual complexity-inspired criticisms of purely logical systems.] But whatever achievement you make, it's just one particular design that may prove effective in some set of domains. And there's the rub - the fact that your design is at least partially static will limit its applicability in some set of domains. I make this argument more completely here: http://www.machineslikeus.com/cms/news/design-bad-or-why-artificial-intelligence-needs-artificial-life or http://tinyurl.com/3coavb If you design a robot, you limit its degrees of freedom. And there will be environments it cannot get around in. By contrast, if you have a design that is capable of changing itself (even if that means from generation to generation), then creative configurations can be discovered. The same basic idea works in the mental arena as well. If you specify the mental machinery, there will be environments it cannot get around in, so to speak. There will be important ways in which it is unable to adapt. You are limiting your design by your own intelligence, which though considerable, is no match for the creativity manifest in a single biological cell. Terren --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com -- Ben Goertzel, PhD CEO, Novamente LLC and Biomind LLC Director of Research, SIAI [EMAIL PROTECTED] Nothing will ever be attempted if all possible objections must be first overcome - Dr Samuel Johnson --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
On Mon, Jun 30, 2008 at 8:07 AM, Terren Suydam [EMAIL PROTECTED] wrote: By the way, just wanted to point out a beautifully simple example - perhaps the simplest - of an irreducibility in complex systems. Individual molecular interactions are symmetric in time, they work the same forwards and backwards. Yet diffusion, which is nothing more than the aggregate of molecular interactions, is asymmetric. Figure that one out. This is just statistical mechanics. The interesting thing is that we make an opportunistic assumption, that any colliding particles are independent before collision (this introduces the time arrow), which is then empirically confirmed by the fact that derived properties agree with the phenomenological theory of entropy. P.S. The biggest issue that spoiled my joy of reading Permutation City is that you cannot simulate dynamic systems ( = solve numerically differential equations) out-of-order, you need to know time t to compute time t+1 (or, alternatively, you need to know t+2), the same goes for space, I presume you need to know x-1,x,x+1 to compute the next-step x. --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
On Mon, Jun 30, 2008 at 8:31 AM, Lukasz Stafiniak [EMAIL PROTECTED] wrote: P.S. The biggest issue that spoiled my joy of reading Permutation City is that you cannot simulate dynamic systems ( = solve numerically differential equations) out-of-order, you need to know time t to compute time t+1 (or, alternatively, you need to know t+2) Yes... the same goes for space, I presume you need to know x-1,x,x+1 to compute the next-step x. No, x+1 is not a function of x. That's the _definition_ of time: a dimension in which t+1 is a function of t. --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Ben, I agree, an evolved design has limits too, but the key difference between a contrived design and one that is allowed to evolve is that the evolved critter's intelligence is grounded in the context of its own 'experience', whereas the contrived one's intelligence is grounded in the experience of its creator, and subject to the limitations built into that conception of intelligence. For example, we really have no idea how we arrive at spontaneous insights (in the shower, for example). A chess master suddenly sees the game-winning move. We can be fairly certain that often, these insights are not the product of logical analysis. So if our conception of intelligence fails to explain these important aspects, our designs based on those conceptions will fail to exhibit them. An evolved intelligence, on the other hand, is not limited in this way, and has the potential to exhibit intelligence in ways we're not capable of comprehending. [btw, I'm using the scare quotes around the word experience as it applies to AGI because it's a controversial word and I hope to convey the basic idea about experience without getting into technical details about it. I can get into that, if anyone thinks it necessary, just didn't want to get bogged down.] Furthermore, there are deeper epistemological issues with the difference between design and self-organization that get into the notion of autonomy as well (i.e., designs lack autonomy to the degree they are specified), but I'll save that for when I feel like putting everyone to sleep :-] Terren PS. As an aside, I believe spontaneous insight is likely to be an example of self-organized criticality, which is a description of the behavior of earthquakes, avalanches, and the punctuated equilibrium model of evolution. Which is to say, a sudden insight is like an avalanche of mental transformations, triggered by some minor event but the result of a build-up of dynamic tension. Self-organized criticality is explained by the late Per Bak in _How Nature Works_, a short, excellent read and an brilliant example of scientific and mathematical progress in the realm of complexity. --- On Mon, 6/30/08, Ben Goertzel [EMAIL PROTECTED] wrote: I agree that all designed systems have limitations, but I also suggest that all evolved systems have limitations. This is just the no free lunch theorem -- in order to perform better than random search at certain optimization tasks, a system needs to have some biases built in, and these biases will cause it to work WORSE than random search on some other optimization tasks. No AGI based on finite resources will ever be **truly** general, be it an engineered or evolved systems Evolved systems are far from being beyond running into dead ends ... their adaptability is far from infinite ... the evolutionary process itself may be endlessly creative, but in that sense so may be the self-modifying process of an engineered AGI ... -- Ben G On Mon, Jun 30, 2008 at 3:17 AM, Terren Suydam [EMAIL PROTECTED] wrote: --- On Mon, 6/30/08, Ben Goertzel [EMAIL PROTECTED] wrote: but I don't agree that predicting **which** AGI designs can lead to the emergent properties corresponding to general intelligence, is pragmatically impossible to do in an analytical and rational way ... OK, I grant you that you may be able to do that. I believe that we can be extremely clever in this regard. An example of that is an implementation of a Turing Machine within the Game of Life: http://rendell-attic.org/gol/tm.htm What a beautiful construction. But it's completely contrived. What you're suggesting is equivalent, because your design is contrived by your own intelligence. [I understand that within the Novamente idea is room for non-deterministic (for practical purposes) behavior, so it doesn't suffer from the usual complexity-inspired criticisms of purely logical systems.] But whatever achievement you make, it's just one particular design that may prove effective in some set of domains. And there's the rub - the fact that your design is at least partially static will limit its applicability in some set of domains. I make this argument more completely here: http://www.machineslikeus.com/cms/news/design-bad-or-why-artificial-intelligence-needs-artificial-life or http://tinyurl.com/3coavb If you design a robot, you limit its degrees of freedom. And there will be environments it cannot get around in. By contrast, if you have a design that is capable of changing itself (even if that means from generation to generation), then creative configurations can be discovered. The same basic idea works in the mental arena as well. If you specify the mental machinery, there will be environments it cannot get around in, so to speak. There will be important ways in which it is unable to adapt. You are limiting your design by your own intelligence, which though considerable, is
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
As far as I can tell, all you've done is give the irreducibility a name: statistical mechanics. You haven't explained how the arrow of time emerges from the local level to the global. Or, maybe I just don't understand it... can you dumb it down for me? Terren --- On Mon, 6/30/08, Lukasz Stafiniak [EMAIL PROTECTED] wrote: [EMAIL PROTECTED] wrote: By the way, just wanted to point out a beautifully simple example - perhaps the simplest - of an irreducibility in complex systems. Individual molecular interactions are symmetric in time, they work the same forwards and backwards. Yet diffusion, which is nothing more than the aggregate of molecular interactions, is asymmetric. Figure that one out. This is just statistical mechanics. The interesting thing is that we make an opportunistic assumption, that any colliding particles are independent before collision (this introduces the time arrow), which is then empirically confirmed by the fact that derived properties agree with the phenomenological theory of entropy. P.S. The biggest issue that spoiled my joy of reading Permutation City is that you cannot simulate dynamic systems ( = solve numerically differential equations) out-of-order, you need to know time t to compute time t+1 (or, alternatively, you need to know t+2), the same goes for space, I presume you need to know x-1,x,x+1 to compute the next-step x. --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
2008/6/30 Terren Suydam [EMAIL PROTECTED]: Ben, I agree, an evolved design has limits too, but the key difference between a contrived design and one that is allowed to evolve is that the evolved critter's intelligence is grounded in the context of its own 'experience', whereas the contrived one's intelligence is grounded in the experience of its creator, and subject to the limitations built into that conception of intelligence. For example, we really have no idea how we arrive at spontaneous insights (in the shower, for example). A chess master suddenly sees the game-winning move. We can be fairly certain that often, these insights are not the product of logical analysis. So if our conception of intelligence fails to explain these important aspects, our designs based on those conceptions will fail to exhibit them. An evolved intelligence, on the other hand, is not limited in this way, and has the potential to exhibit intelligence in ways we're not capable of comprehending. I'm seeking to do something half way between what you suggest (from bacterial systems to human alife) and AI. I'd be curious to know whether you think it would suffer from the same problems. First are we agreed that the von Neumann model of computing has no hidden bias to its problem solving capabilities. It might be able to do some jobs more efficiently than other and need lots of memory to do others but it is not particularly suited to learning chess or running down a gazelle. Which means it can be reprogrammed to do either. However it has no guide to what it should be doing, so can become virus infested or subverted. It has a purpose but we can't explicitly define it. So let us try and put in the most minimal guide that we can so we don't give it a specific goal, just a tendency to favour certain activities or programs. How to do this? Form and economy based on reinforcement signals, those that get more reinforcement signals can outbid the others for control of system resources. This is obviously reminiscent of tierra and a million and one other alife system. The difference being is that I want the whole system to exhibit intelligence. Any form of variation is allowed, from random to getting in programs from the outside. It should be able to change the whole from the OS level up based on the variation. I agree that we want the systems we make to be free of our design constraints long term, that is eventually correct all the errors and oversimplifications or gaps we left. But I don't see the need to go all the way back to bacteria. Even then you would need to design the system correctly in terms of chemical concentrations. I think both would count as the passive approach* to helping solve the problem, yours is more indirect than is needed I think. Will Pearson * http://www.mail-archive.com/agi@v2.listbox.com/msg11399.html --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
On Mon, Jun 30, 2008 at 10:34 PM, William Pearson [EMAIL PROTECTED] wrote: I'm seeking to do something half way between what you suggest (from bacterial systems to human alife) and AI. I'd be curious to know whether you think it would suffer from the same problems. First are we agreed that the von Neumann model of computing has no hidden bias to its problem solving capabilities. It might be able to do some jobs more efficiently than other and need lots of memory to do others but it is not particularly suited to learning chess or running down a gazelle. Which means it can be reprogrammed to do either. However it has no guide to what it should be doing, so can become virus infested or subverted. It has a purpose but we can't explicitly define it. So let us try and put in the most minimal guide that we can so we don't give it a specific goal, just a tendency to favour certain activities or programs. It is a wrong level of organization: computing hardware is the physics of computation, it isn't meant to implement specific algorithms, so I don't quite see what you are arguing. How to do this? Form and economy based on reinforcement signals, those that get more reinforcement signals can outbid the others for control of system resources. Where do reinforcement signals come from? What does this specification improve over natural evolution that needed billions of years to get here (that is, why do you expect any results in the forseable future)? This is obviously reminiscent of tierra and a million and one other alife system. The difference being is that I want the whole system to exhibit intelligence. Any form of variation is allowed, from random to getting in programs from the outside. It should be able to change the whole from the OS level up based on the variation. What is your meaning of `intelligence'? I now see it as merely the efficiency of optimization process that drives the environment towards higher utility, according to whatever criterion (reinforcement, in your case). In this view, how does I'll do the same, but with intelligence differ from I'll do the same, but better? -- Vladimir Nesov [EMAIL PROTECTED] http://causalityrelay.wordpress.com/ --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Hi William, A Von Neumann computer is just a machine. It's only purpose is to compute. When you get into higher-level purpose, you have to go up a level to the stuff being computed. Even then, the purpose is in the mind of the programmer. The only way to talk coherently about purpose within the computation is to simulate self-organized, embodied systems. And I applaud your intuition to make the whole system intelligent. One of my biggest criticisms of traditional AI philosophy is over-emphasis on the agent. Indeed, the ideal simulation, in my mind, is one in which the boundary between agent and environment is blurry. In nature, for example, at low-enough levels of description it is impossible to find a boundary between the two, because the entities at that level are freely exchanged. You are right that starting with bacteria is too indirect, if your goal is to achieve AGI in something like decades. It would certainly take an enormous amount of time and computation to get from there to human-level AI and beyond, perhaps a hundred years or more. But you're asking, aren't there shortcuts we can take that don't limit the field of potential intelligence in important ways. For example, starting with bacteria means we have to let multi-cellular organisms evolve on their own in a virtual geometry. That project alone is an enormous challenge. So let's skip it and go right to the multi-cellular design. The trouble is, our design of the multi-cellular organism is limiting. Alternative designs become impossible. The question at that point is, are we excluding any important possibilities for intelligence if we build in our assumptions about what is necessary to support it, on a low-level basis. In what ways is our designed brain leaving out some key to adapting to unforeseen domains? One of the basic threads of scientific progress is the ceaseless denigration of the idea that there is something special about humans. Pretending that we can solve AGI by mimicking top-down high-level human reasoning is another example of that kind of hubris, and eventually, that idea will fall too. Terren --- On Mon, 6/30/08, William Pearson [EMAIL PROTECTED] wrote: Ben, I agree, an evolved design has limits too, but the key difference between a contrived design and one that is allowed to evolve is that the evolved critter's intelligence is grounded in the context of its own 'experience', whereas the contrived one's intelligence is grounded in the experience of its creator, and subject to the limitations built into that conception of intelligence. For example, we really have no idea how we arrive at spontaneous insights (in the shower, for example). A chess master suddenly sees the game-winning move. We can be fairly certain that often, these insights are not the product of logical analysis. So if our conception of intelligence fails to explain these important aspects, our designs based on those conceptions will fail to exhibit them. An evolved intelligence, on the other hand, is not limited in this way, and has the potential to exhibit intelligence in ways we're not capable of comprehending. I'm seeking to do something half way between what you suggest (from bacterial systems to human alife) and AI. I'd be curious to know whether you think it would suffer from the same problems. First are we agreed that the von Neumann model of computing has no hidden bias to its problem solving capabilities. It might be able to do some jobs more efficiently than other and need lots of memory to do others but it is not particularly suited to learning chess or running down a gazelle. Which means it can be reprogrammed to do either. However it has no guide to what it should be doing, so can become virus infested or subverted. It has a purpose but we can't explicitly define it. So let us try and put in the most minimal guide that we can so we don't give it a specific goal, just a tendency to favour certain activities or programs. How to do this? Form and economy based on reinforcement signals, those that get more reinforcement signals can outbid the others for control of system resources. This is obviously reminiscent of tierra and a million and one other alife system. The difference being is that I want the whole system to exhibit intelligence. Any form of variation is allowed, from random to getting in programs from the outside. It should be able to change the whole from the OS level up based on the variation. I agree that we want the systems we make to be free of our design constraints long term, that is eventually correct all the errors and oversimplifications or gaps we left. But I don't see the need to go all the way back to bacteria. Even then you would need to design the system correctly in terms of chemical concentrations. I think both would count as the passive approach* to helping solve the problem, yours is more
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Terren:One of the basic threads of scientific progress is the ceaseless denigration of the idea that there is something special about humans Not quite so. There is a great deal of exceptionalism in science - hence evolutionary psychology actually only deals with human evolution. If there were a true all-species evolutionary psychology, that really did look at the evolution of the mind through all species, you wouldn't get what will come to be seen as the mechanistic absurdity of trying to create an AGI starting at human level. To twist a favourite analogy of AGI-ers, that's like trying to start mechanical invention at the airplane stage, and jump the billions of steps from rock tools and wheels on - or trying to invent a computer before electricity has been discovered. --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Hello Terren A Von Neumann computer is just a machine. It's only purpose is to compute. When you get into higher-level purpose, you have to go up a level to the stuff being computed. Even then, the purpose is in the mind of the programmer. What I don't see is why your simulation gets away from this, where as my architecture doesn't. Read the linked post in the previous message, if you want to understand more about the philosophy of the system. The only way to talk coherently about purpose within the computation is to simulate self-organized, embodied systems. I don't think you are quite getting my system. If you had a bunch of programs that did the following 1) created new programs, by trial and error and taking statistics of variables or getting arbitrary code from the outside. 2) communicated with each other to try and find programs that perform services they need. 3) Bid for computer resources, if a program loses its memory resources it is selected against, in a way. Would this be sufficiently self-organised? If not, why not? And the computer programs would be as embodied as your virtual creatures. They would just be embodied within a tacit economy, rather than an artificial chemistry. And I applaud your intuition to make the whole system intelligent. One of my biggest criticisms of traditional AI philosophy is over-emphasis on the agent. Indeed, the ideal simulation, in my mind, is one in which the boundary between agent and environment is blurry. In nature, for example, at low-enough levels of description it is impossible to find a boundary between the two, because the entities at that level are freely exchanged. You are right that starting with bacteria is too indirect, if your goal is to achieve AGI in something like decades. It would certainly take an enormous amount of time and computation to get from there to human-level AI and beyond, perhaps a hundred years or more. But you're asking, aren't there shortcuts we can take that don't limit the field of potential intelligence in important ways. If you take this attitude you would have to ask yourself whether implementing your simulation on a classical computer is not cutting off the ability to create intelligence. Perhaps quantum affects are important in whether a system can produce intelligence. Protein folding probably wouldn't be the same. You have to at some point simplify. I'm going to have my system have as many degrees of freedom to vary as a stored program computer (or as near as I can make it). Whilst having the internal programs self-organise and vary in ways that would make a normal stored program computer become unstable. Any simulations you do on a computer cannot have any more degrees of freedom. For example, starting with bacteria means we have to let multi-cellular organisms evolve on their own in a virtual geometry. That project alone is an enormous challenge. So let's skip it and go right to the multi-cellular design. The trouble is, our design of the multi-cellular organism is limiting. Alternative designs become impossible. What do you mean by design here? Do you mean an abstract multicellular cell model or do you mean design as in what Tom Ray (you do know Tierra right, I can use this as a common language?) did with his first self replicator, by creating an artificial genome. I can see problems with the first in restricting degrees of freedom, but the second, the degrees of freedom are still there to be acted on by the pressures of variation within the system. Even though Tom Ray built a certain type of replicator, they still managed to replicate in other ways, the one I can remember is stealing other peoples replication machinery as parasites. Lets say you started with an artificial chemistry. You could then design within that chemistry a replicator, then test that replicator. See if the variation is working okay. Then design a multicellular variant, by changing its genome. It could still slip back to single cellularity and find a different way to multicellularity. The degrees of freedom do not go away the second a human starts to design something (else genetically modified foods would not be such a thorny issue), you just got to allow the forces of variation to be able to act upon them. The question at that point is, are we excluding any important possibilities for intelligence if we build in our assumptions about what is necessary to support it, on a low-level basis. In what ways is our designed brain leaving out some key to adapting to unforeseen domains? Just apply a patch :P Or have an architecture that is capable of supporting a self-patching system. I have no fixed design for an AI myself. Intelligence means winning, winning requires flexibility. One of the basic threads of scientific progress is the ceaseless denigration of the idea that there is something special about humans. Pretending that we can solve AGI by mimicking top-down high-level human
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Hi Mike, Evidently I didn't communicate that so clearly because I agree with you 100%. Terren --- On Mon, 6/30/08, Mike Tintner [EMAIL PROTECTED] wrote: Terren:One of the basic threads of scientific progress is the ceaseless denigration of the idea that there is something special about humans Not quite so. There is a great deal of exceptionalism in science - hence evolutionary psychology actually only deals with human evolution. If there were a true all-species evolutionary psychology, that really did look at the evolution of the mind through all species, you wouldn't get what will come to be seen as the mechanistic absurdity of trying to create an AGI starting at human level. To twist a favourite analogy of AGI-ers, that's like trying to start mechanical invention at the airplane stage, and jump the billions of steps from rock tools and wheels on - or trying to invent a computer before electricity has been discovered. --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
2008/6/30 Vladimir Nesov [EMAIL PROTECTED]: On Mon, Jun 30, 2008 at 10:34 PM, William Pearson [EMAIL PROTECTED] wrote: I'm seeking to do something half way between what you suggest (from bacterial systems to human alife) and AI. I'd be curious to know whether you think it would suffer from the same problems. First are we agreed that the von Neumann model of computing has no hidden bias to its problem solving capabilities. It might be able to do some jobs more efficiently than other and need lots of memory to do others but it is not particularly suited to learning chess or running down a gazelle. Which means it can be reprogrammed to do either. However it has no guide to what it should be doing, so can become virus infested or subverted. It has a purpose but we can't explicitly define it. So let us try and put in the most minimal guide that we can so we don't give it a specific goal, just a tendency to favour certain activities or programs. It is a wrong level of organization: computing hardware is the physics of computation, it isn't meant to implement specific algorithms, so I don't quite see what you are arguing. I'm not implementing a specific algorithm I am controlling how resources are allocated. Currently architecture does whatever the kernel says, from memory allocation to irq allocation. Instead of this my architecture would allow any program to bid credit for a resource. The one that bids the most wins and spends its credit. Certain resources like output memory space, (i.e if the program is controlling the display or an arm or something) allow the program to specify a bank, and give the program income. A bank is a special variable that can't be edited by programs normally but can be spent. The bank of an outputing program will be given credit depending upon how well the system as whole is performing . If it is doing well the amount of credit it gets would be above average, poorly it would be below. After a certain time the resources will need to be bid for again. So credit is coming into the system and continually being sunk. The system will be seeded with programs that can perform rudimentarily well. E.g. you will have programs that know how to deal with visual input and they will bid for the video camera interupt. They will then sell their services for credit (so that they can bid for the interrupt again), to a program that correlates visual and auditory responses. Who sell their services to a high level planning module etc, on down to the arm that actually gets the credit. All these modules are subject to change and re-evaluation. They merely suggest one possible way for it to be used. It is supposed to be ultimately flexible. You could seed it with a self-replicating neural simulator that tried to hook its inputs and outputs up to other neurons. Neurons would die out if they couldn't find anything to do. How to do this? Form and economy based on reinforcement signals, those that get more reinforcement signals can outbid the others for control of system resources. Where do reinforcement signals come from? What does this specification improve over natural evolution that needed billions of years to get here (that is, why do you expect any results in the forseable future)? Most of the internals are programmed by humans, and they can be arbitrarily complex. The feedback comes from a human, or from a utility function although those are harder to define. The architecture simply doesn't restrict the degrees of freedom that the programs inside it can explore. This is obviously reminiscent of tierra and a million and one other alife system. The difference being is that I want the whole system to exhibit intelligence. Any form of variation is allowed, from random to getting in programs from the outside. It should be able to change the whole from the OS level up based on the variation. What is your meaning of `intelligence'? I now see it as merely the efficiency of optimization process that drives the environment towards higher utility, according to whatever criterion (reinforcement, in your case). In this view, how does I'll do the same, but with intelligence differ from I'll do the same, but better? Terran's artificial chemistry as whole could not be said to have a goal. Or to put it another way applying the intentional stance to it probably wouldn't help you predict what it did next. Applying the intentional stance to what my system does should help you predict what it does. This means he needs to use a bunch more resources to get a singular useful system. Also the system might not do what he wants, but I don't think he minds about that. I'm allowing humans to design everything, just allowing the very low level to vary. Is this clearer? Will Pearson --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription:
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
On Tue, Jul 1, 2008 at 1:31 AM, William Pearson [EMAIL PROTECTED] wrote: 2008/6/30 Vladimir Nesov [EMAIL PROTECTED]: It is a wrong level of organization: computing hardware is the physics of computation, it isn't meant to implement specific algorithms, so I don't quite see what you are arguing. I'm not implementing a specific algorithm I am controlling how resources are allocated. Currently architecture does whatever the kernel says, from memory allocation to irq allocation. Instead of this my architecture would allow any program to bid credit for a resource. The one that bids the most wins and spends its credit. Certain resources like output memory space, (i.e if the program is controlling the display or an arm or something) allow the program to specify a bank, and give the program income. A bank is a special variable that can't be edited by programs normally but can be spent. The bank of an outputing program will be given credit depending upon how well the system as whole is performing . If it is doing well the amount of credit it gets would be above average, poorly it would be below. After a certain time the resources will need to be bid for again. So credit is coming into the system and continually being sunk. The system will be seeded with programs that can perform rudimentarily well. E.g. you will have programs that know how to deal with visual input and they will bid for the video camera interupt. They will then sell their services for credit (so that they can bid for the interrupt again), to a program that correlates visual and auditory responses. Who sell their services to a high level planning module etc, on down to the arm that actually gets the credit. All these modules are subject to change and re-evaluation. They merely suggest one possible way for it to be used. It is supposed to be ultimately flexible. You could seed it with a self-replicating neural simulator that tried to hook its inputs and outputs up to other neurons. Neurons would die out if they couldn't find anything to do. Well, yes, you implement some functionality, but why would you contrast it with underlying levels (hardware, OS)? Like Java virtual machine, your system is a platform, and it does some things not handled by lower levels, or, in this case, by any superficially analogous platforms. How to do this? Form and economy based on reinforcement signals, those that get more reinforcement signals can outbid the others for control of system resources. Where do reinforcement signals come from? What does this specification improve over natural evolution that needed billions of years to get here (that is, why do you expect any results in the forseable future)? Most of the internals are programmed by humans, and they can be arbitrarily complex. The feedback comes from a human, or from a utility function although those are harder to define. The architecture simply doesn't restrict the degrees of freedom that the programs inside it can explore. If internals are programmed by humans, why do you need automatic system to assess them? It would be useful if you needed to construct and test some kind of combination/setting automatically, but not if you just test manually-programmed systems. How does the assessment platform help in improving/accelerating the research? This is obviously reminiscent of tierra and a million and one other alife system. The difference being is that I want the whole system to exhibit intelligence. Any form of variation is allowed, from random to getting in programs from the outside. It should be able to change the whole from the OS level up based on the variation. What is your meaning of `intelligence'? I now see it as merely the efficiency of optimization process that drives the environment towards higher utility, according to whatever criterion (reinforcement, in your case). In this view, how does I'll do the same, but with intelligence differ from I'll do the same, but better? Terran's artificial chemistry as whole could not be said to have a goal. Or to put it another way applying the intentional stance to it probably wouldn't help you predict what it did next. Applying the intentional stance to what my system does should help you predict what it does. What is `intentional stance'? Intentional stance of what? What is it good for? This means he needs to use a bunch more resources to get a singular useful system. Also the system might not do what he wants, but I don't think he minds about that. I'm allowing humans to design everything, just allowing the very low level to vary. Is this clearer? What do you mean by varying low level, especially in human-designed systems? -- Vladimir Nesov [EMAIL PROTECTED] http://causalityrelay.wordpress.com/ --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
2008/6/30 Vladimir Nesov [EMAIL PROTECTED]: On Tue, Jul 1, 2008 at 1:31 AM, William Pearson [EMAIL PROTECTED] wrote: 2008/6/30 Vladimir Nesov [EMAIL PROTECTED]: It is a wrong level of organization: computing hardware is the physics of computation, it isn't meant to implement specific algorithms, so I don't quite see what you are arguing. I'm not implementing a specific algorithm I am controlling how resources are allocated. Currently architecture does whatever the kernel says, from memory allocation to irq allocation. Instead of this my architecture would allow any program to bid credit for a resource. The one that bids the most wins and spends its credit. Certain resources like output memory space, (i.e if the program is controlling the display or an arm or something) allow the program to specify a bank, and give the program income. A bank is a special variable that can't be edited by programs normally but can be spent. The bank of an outputing program will be given credit depending upon how well the system as whole is performing . If it is doing well the amount of credit it gets would be above average, poorly it would be below. After a certain time the resources will need to be bid for again. So credit is coming into the system and continually being sunk. The system will be seeded with programs that can perform rudimentarily well. E.g. you will have programs that know how to deal with visual input and they will bid for the video camera interupt. They will then sell their services for credit (so that they can bid for the interrupt again), to a program that correlates visual and auditory responses. Who sell their services to a high level planning module etc, on down to the arm that actually gets the credit. All these modules are subject to change and re-evaluation. They merely suggest one possible way for it to be used. It is supposed to be ultimately flexible. You could seed it with a self-replicating neural simulator that tried to hook its inputs and outputs up to other neurons. Neurons would die out if they couldn't find anything to do. Well, yes, you implement some functionality, but why would you contrast it with underlying levels (hardware, OS)? Like Java virtual machine, your system is a platform, and it does some things not handled by lower levels, or, in this case, by any superficially analogous platforms. Because I want it done in silicon at some stage. It is also assumed to be the whole system, that is no other significant programs on it. Machines that run lisp natively have been made, this makes the most sense as the whole computer. Rather than as a component. How to do this? Form and economy based on reinforcement signals, those that get more reinforcement signals can outbid the others for control of system resources. Where do reinforcement signals come from? What does this specification improve over natural evolution that needed billions of years to get here (that is, why do you expect any results in the forseable future)? Most of the internals are programmed by humans, and they can be arbitrarily complex. The feedback comes from a human, or from a utility function although those are harder to define. The architecture simply doesn't restrict the degrees of freedom that the programs inside it can explore. If internals are programmed by humans, why do you need automatic system to assess them? It would be useful if you needed to construct and test some kind of combination/setting automatically, but not if you just test manually-programmed systems. How does the assessment platform help in improving/accelerating the research? Because to be interesting the human specified programs need to be autogenous, as in Josh Storr Hall's terminology, which means self-building. Capable of altering the stuff they are made of. In this case machine code equivalent. So you need the human to assess the improvements the system makes, for whatever purpose the human wants the system to perform. This is obviously reminiscent of tierra and a million and one other alife system. The difference being is that I want the whole system to exhibit intelligence. Any form of variation is allowed, from random to getting in programs from the outside. It should be able to change the whole from the OS level up based on the variation. What is your meaning of `intelligence'? I now see it as merely the efficiency of optimization process that drives the environment towards higher utility, according to whatever criterion (reinforcement, in your case). In this view, how does I'll do the same, but with intelligence differ from I'll do the same, but better? Terran's artificial chemistry as whole could not be said to have a goal. Or to put it another way applying the intentional stance to it probably wouldn't help you predict what it did next. Applying the intentional stance to what my system does should help you predict what it does. What is
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
I wrote a book about the emergence of spontaneous creativity from underlying complex dynamics. It was published in 1997 with the title From Complexity to Creativity. Some of the material is dated but I still believe the basic ideas make sense. Some of the main ideas were reviewed in The Hidden Pattern (2006). I don't have time to review the ideas right now (I'm in an airport during a flight change doing a quick email check) but suffice to say that I did put a lot of thought and analysis into how spontaneous creativity emerges from complex cognitive systems. So have others. It is not a total mystery, as mysterious as the experience can seem subjectively. -- Ben G On Mon, Jun 30, 2008 at 1:32 PM, Terren Suydam [EMAIL PROTECTED] wrote: Ben, I agree, an evolved design has limits too, but the key difference between a contrived design and one that is allowed to evolve is that the evolved critter's intelligence is grounded in the context of its own 'experience', whereas the contrived one's intelligence is grounded in the experience of its creator, and subject to the limitations built into that conception of intelligence. For example, we really have no idea how we arrive at spontaneous insights (in the shower, for example). A chess master suddenly sees the game-winning move. We can be fairly certain that often, these insights are not the product of logical analysis. So if our conception of intelligence fails to explain these important aspects, our designs based on those conceptions will fail to exhibit them. An evolved intelligence, on the other hand, is not limited in this way, and has the potential to exhibit intelligence in ways we're not capable of comprehending. [btw, I'm using the scare quotes around the word experience as it applies to AGI because it's a controversial word and I hope to convey the basic idea about experience without getting into technical details about it. I can get into that, if anyone thinks it necessary, just didn't want to get bogged down.] Furthermore, there are deeper epistemological issues with the difference between design and self-organization that get into the notion of autonomy as well (i.e., designs lack autonomy to the degree they are specified), but I'll save that for when I feel like putting everyone to sleep :-] Terren PS. As an aside, I believe spontaneous insight is likely to be an example of self-organized criticality, which is a description of the behavior of earthquakes, avalanches, and the punctuated equilibrium model of evolution. Which is to say, a sudden insight is like an avalanche of mental transformations, triggered by some minor event but the result of a build-up of dynamic tension. Self-organized criticality is explained by the late Per Bak in _How Nature Works_, a short, excellent read and an brilliant example of scientific and mathematical progress in the realm of complexity. --- On Mon, 6/30/08, Ben Goertzel [EMAIL PROTECTED] wrote: I agree that all designed systems have limitations, but I also suggest that all evolved systems have limitations. This is just the no free lunch theorem -- in order to perform better than random search at certain optimization tasks, a system needs to have some biases built in, and these biases will cause it to work WORSE than random search on some other optimization tasks. No AGI based on finite resources will ever be **truly** general, be it an engineered or evolved systems Evolved systems are far from being beyond running into dead ends ... their adaptability is far from infinite ... the evolutionary process itself may be endlessly creative, but in that sense so may be the self-modifying process of an engineered AGI ... -- Ben G On Mon, Jun 30, 2008 at 3:17 AM, Terren Suydam [EMAIL PROTECTED] wrote: --- On Mon, 6/30/08, Ben Goertzel [EMAIL PROTECTED] wrote: but I don't agree that predicting **which** AGI designs can lead to the emergent properties corresponding to general intelligence, is pragmatically impossible to do in an analytical and rational way ... OK, I grant you that you may be able to do that. I believe that we can be extremely clever in this regard. An example of that is an implementation of a Turing Machine within the Game of Life: http://rendell-attic.org/gol/tm.htm What a beautiful construction. But it's completely contrived. What you're suggesting is equivalent, because your design is contrived by your own intelligence. [I understand that within the Novamente idea is room for non-deterministic (for practical purposes) behavior, so it doesn't suffer from the usual complexity-inspired criticisms of purely logical systems.] But whatever achievement you make, it's just one particular design that may prove effective in some set of domains. And there's the rub - the fact that your design is at least partially static will limit its applicability in some set of domains. I
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Hi Will, --- On Mon, 6/30/08, William Pearson [EMAIL PROTECTED] wrote: The only way to talk coherently about purpose within the computation is to simulate self-organized, embodied systems. I don't think you are quite getting my system. If you had a bunch of programs that did the following 1) created new programs, by trial and error and taking statistics of variables or getting arbitrary code from the outside. 2) communicated with each other to try and find programs that perform services they need. 3) Bid for computer resources, if a program loses its memory resources it is selected against, in a way. Would this be sufficiently self-organised? If not, why not? And the computer programs would be as embodied as your virtual creatures. They would just be embodied within a tacit economy, rather than an artificial chemistry. It boils down to your answer to the question: how are the resources ultimately allocated to the programs? If you're the one specifying it, via some heuristic or rule, then the purpose is driven by you. If resource allocation is handled by some self-organizing method (this wasn't clear in the article you provided), then I'd say that the system's purpose is self-defined. As for embodiment, my question is, how do your programs receive input? Embodiment, as I define it, requires that inputs are merely reflections of state variables, and not even labeled in any way... i.e. we can't pre-define ontologies. The embodied entity starts from the most unstructured state possible and self-structures whatever inputs it receives. That said, you may very well be doing that and be creating embodied programs in this way... if so, that's cool because I hadn't considered that possibility and I'll be interested to see how you fare. You are right that starting with bacteria is too indirect, if your goal is to achieve AGI in something like decades. It would certainly take an enormous amount of time and computation to get from there to human-level AI and beyond, perhaps a hundred years or more. But you're asking, aren't there shortcuts we can take that don't limit the field of potential intelligence in important ways. If you take this attitude you would have to ask yourself whether implementing your simulation on a classical computer is not cutting off the ability to create intelligence. Perhaps quantum affects are important in whether a system can produce intelligence. Protein folding probably wouldn't be the same. Computation per se has little to do with the potential to create intelligent systems. Computation is only a framework that supports the simulation of virtual environments, in which intelligence may emerge. You could in principle build that computer out of tinker toys, or as an implementation of a Turing machine in Conway's Game of Life. The substrate doesn't matter, so long as it can compute. As for quantum effects, it's possible there's something there with respect to protein folding, probable even. But I strongly distrust attempts to locate the non-deterministic behavior required of autonomous systems in the domain of quantum uncertainty. Every phenomenon above the scale of molecular dynamics is far too large to be impacted by anything but statistical behaviors. Individual quantal events lose all practical meaning at that level. Because intelligence, in my estimation, is at least partially dependent on global notions of emergence and complexity, quantum effects contribute absolutely nothing to my model. You have to at some point simplify. I'm going to have my system have as many degrees of freedom to vary as a stored program computer (or as near as I can make it). Whilst having the internal programs self-organise and vary in ways that would make a normal stored program computer become unstable. Any simulations you do on a computer cannot have any more degrees of freedom. I disagree, but would like to see your response to the above before diving into such esoterica. For example, starting with bacteria means we have to let multi-cellular organisms evolve on their own in a virtual geometry. That project alone is an enormous challenge. So let's skip it and go right to the multi-cellular design. The trouble is, our design of the multi-cellular organism is limiting. Alternative designs become impossible. What do you mean by design here? Do you mean an abstract multicellular cell model or do you mean design as in what Tom Ray (you do know Tierra right, I can use this as a common language?) did with his first self replicator, by creating an artificial genome. I can see problems with the first in restricting degrees of freedom, but the second, the degrees of freedom are still there to be acted on by the pressures of variation within the system. Even though Tom Ray built a certain type of replicator, they still managed to replicate in other ways, the one I can remember is stealing other
RE: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Could you say that it takes a complex system to know a complex system? If an AGI is going to try to say predict the weather, it doesn't have infinite cpu cycles to simulate so it'll have to come up with something better. Sure it can build a probabilistic historical model but that is kind of cheating. So for it to emulate the weather, I think, or to semi-understand it there has to be some complex systems activity going on there in its cognition. No? I'm not sure that this what Richard is taking about but an AGI is going to bump into complex systems all over the place. Also it will encounter what seems to be complex and later on it may determine that it is not. And perhaps, a key component in the cognition engine in order for it to understand complexity differentials in systems from a relationist standpoint it would need some sort of complexity .. not a comparator but a...sort of harmonic leverage. Can't think of the right words Either way this complexity thing is getting rather annoying because on one hand you think it can drasticly enhance an AGI and is required and on the other hand you think it is unnecessary - I'm not talking about creativity or thought emergence or similar but complexity as integral component in a computational cognition system. John --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Ben, Be that as it may, spontaneous insight was just one example of an aspect of human intelligence that's not well understood. I'll give you another one that is more difficult to theorize about - I assume you've heard of the savant Daniel Tammet who is able to do amazing feats of computation and memory? He appears to be doing these things in a way that does not involve algorithmic computation, as vetted by Vilayanur Ramachandran. Tammet is clearly one of the most intelligent people on the planet by many measures, and he does it in ways we don't understand. This isn't an invitation to theorize about Tammet's cognitive machinery, as interesting as that exercise might be. It's to make the larger point that we may be so immersed in our own conceptualizations of intelligence - particularly because we live in our models and draw on our own experience and introspection to elaborate them - that we may have tunnel vision about the possibilities for better or different models. Or, we may take for granted huge swaths of what makes us so smart, because it's so familiar, or below the radar of our conscious awareness, that it doesn't even occur to us to reflect on it. A perfect example of that is how we acquire language (our first language). Introspection is not available to us there, so all we have is theory. And even when introspection *is* available to us, we may fall prey to the self-deception that is such an integral part of human psychology. In short, claiming that your particular design is capable of AGI is quite a bold claim, because of all the possible pitfalls involved with theorizing about human-level intelligence. Given that the graveyard of AI's history is strewn with the bones of outrageous boasts and predictions, it's too tempting to see Novamente as just the latest in a long lineage. Why do we insist on shooting for the moon, when we still can't even explain the brain of a housefly? One of the best reasons to go with an evolving-design approach is that we're not pretending we're going to get to human-level AI on the first shot. Instead, we gradually build up the complexity of our creations, building on prior successes and milestones. We see the evolution of intelligence as it becomes progressively more complicated. Instead of leaping off a cliff (like Icarus), we climb a mountain (like Sisyphus ;-). Progress is measurable and reflects the graduated spectrum of intelligence, a nuance that has never been fashionable. Terren --- On Mon, 6/30/08, Ben Goertzel [EMAIL PROTECTED] wrote: I wrote a book about the emergence of spontaneous creativity from underlying complex dynamics. It was published in 1997 with the title From Complexity to Creativity. Some of the material is dated but I still believe the basic ideas make sense. Some of the main ideas were reviewed in The Hidden Pattern (2006). I don't have time to review the ideas right now (I'm in an airport during a flight change doing a quick email check) but suffice to say that I did put a lot of thought and analysis into how spontaneous creativity emerges from complex cognitive systems. So have others. It is not a total mystery, as mysterious as the experience can seem subjectively. -- Ben G On Mon, Jun 30, 2008 at 1:32 PM, Terren Suydam [EMAIL PROTECTED] wrote: Ben, I agree, an evolved design has limits too, but the key difference between a contrived design and one that is allowed to evolve is that the evolved critter's intelligence is grounded in the context of its own 'experience', whereas the contrived one's intelligence is grounded in the experience of its creator, and subject to the limitations built into that conception of intelligence. For example, we really have no idea how we arrive at spontaneous insights (in the shower, for example). A chess master suddenly sees the game-winning move. We can be fairly certain that often, these insights are not the product of logical analysis. So if our conception of intelligence fails to explain these important aspects, our designs based on those conceptions will fail to exhibit them. An evolved intelligence, on the other hand, is not limited in this way, and has the potential to exhibit intelligence in ways we're not capable of comprehending. [btw, I'm using the scare quotes around the word experience as it applies to AGI because it's a controversial word and I hope to convey the basic idea about experience without getting into technical details about it. I can get into that, if anyone thinks it necessary, just didn't want to get bogged down.] Furthermore, there are deeper epistemological issues with the difference between design and self-organization that get into the notion of autonomy as well (i.e., designs lack autonomy to the degree they are specified), but I'll save that for when I feel like putting everyone to sleep :-] Terren PS. As an aside, I believe spontaneous insight is likely
Savants and user-interfaces [was Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
2008/6/30 Terren Suydam [EMAIL PROTECTED]: savant I've always theorized that savants can do what they do because they've been able to get direct access to, and train, a fairly small number of neurons in their brain, to accomplish highly specialized (and thus rather unusual) calculations. I'm thinking specifically of Ramanujan, the Hindi mathematician. He appears to have had access to a multiply-add type circuit in his brain, and could do symbolic long division and multiplication as a result -- I base this on studying some of the things he came up with -- after a while, it seems to be clear how he came up with it (even if the feat is clearly not reproducible). In a sense, similar feats are possible by using a modern computer with a good algebra system. Simon Plouffe seems to be a modern-day example of this: he noodles around with his systems, and finds various interesting relationships that would otherwise be obscure/unknown. He does this without any particularly deep or expansive training in math (whence some of his friction with real academics). If Simon could get a computer-algebra chip implanted in his brain, (i.e. with a very, very user-freindly user-interface) so that he could work the algebra system just by thinking about it, I bet his output would resemble that of Ramanujan a whole lot more than it already does -- as it were, he's hobbled by a crappy user interface. Thus, let me theorize: by studying savants with MRI and what-not, we may find a way of getting a much better man-machine interface. That is, currently, electrodes are always implanted in motor neurons (or visual cortex, etc) i.e. in places of the brain with very low levels of abstraction from the real word. It would be interesting to move up the level of abstraction, and I think that studying how savants access the magic circuits in thier brain will open up a method for high-level interfaces to external computing machinery. --linas --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: Savants and user-interfaces [was Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Interesting: is it possible to train yourself to run a specially designed nontrivial inference circuit based on low-base transformations (e.g. binary)? You start by assigning unique symbols to its nodes, train yourself to stably perform associations implementing its junctions, and then assemble it all together by training yourself to generate a problem as a temporal sequence (request), so that it can be handled by the overall circuit, and training to read out the answer and convert it to sequence of e.g. base-10 digits or base-100 words keying pairs of digits (like in mnemonic)? Has anyone heard of this attempted? At least the initial steps look straightforward enough, what kind of obstacles this kind of experiment can run into? On Tue, Jul 1, 2008 at 7:43 AM, Linas Vepstas [EMAIL PROTECTED] wrote: 2008/6/30 Terren Suydam [EMAIL PROTECTED]: savant I've always theorized that savants can do what they do because they've been able to get direct access to, and train, a fairly small number of neurons in their brain, to accomplish highly specialized (and thus rather unusual) calculations. I'm thinking specifically of Ramanujan, the Hindi mathematician. He appears to have had access to a multiply-add type circuit in his brain, and could do symbolic long division and multiplication as a result -- I base this on studying some of the things he came up with -- after a while, it seems to be clear how he came up with it (even if the feat is clearly not reproducible). In a sense, similar feats are possible by using a modern computer with a good algebra system. Simon Plouffe seems to be a modern-day example of this: he noodles around with his systems, and finds various interesting relationships that would otherwise be obscure/unknown. He does this without any particularly deep or expansive training in math (whence some of his friction with real academics). If Simon could get a computer-algebra chip implanted in his brain, (i.e. with a very, very user-freindly user-interface) so that he could work the algebra system just by thinking about it, I bet his output would resemble that of Ramanujan a whole lot more than it already does -- as it were, he's hobbled by a crappy user interface. Thus, let me theorize: by studying savants with MRI and what-not, we may find a way of getting a much better man-machine interface. That is, currently, electrodes are always implanted in motor neurons (or visual cortex, etc) i.e. in places of the brain with very low levels of abstraction from the real word. It would be interesting to move up the level of abstraction, and I think that studying how savants access the magic circuits in thier brain will open up a method for high-level interfaces to external computing machinery. --linas --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com -- Vladimir Nesov [EMAIL PROTECTED] http://causalityrelay.wordpress.com/ --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: Savants and user-interfaces [was Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
2008/6/30 Vladimir Nesov [EMAIL PROTECTED]: Interesting: is it possible to train yourself to run a specially designed nontrivial inference circuit based on low-base transformations (e.g. binary)? Why binary? I once skimmed a biography of Ramanujan, he started multiplying numbers in his head as a pre-teen. I suspect it was grindingly boring, but given the surroundings, might have been the most fun thing he could think of. If you're autistic, then focusing obsessively on some task might be a great way to pass the time, but if you're more or less normal, I doubt you'll get very far with obsessive-compulsive self-training -- and that's the problem, isn't it? --linas --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: Savants and user-interfaces [was Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
On Tue, Jul 1, 2008 at 8:31 AM, Linas Vepstas [EMAIL PROTECTED] wrote: Why binary? I once skimmed a biography of Ramanujan, he started multiplying numbers in his head as a pre-teen. I suspect it was grindingly boring, but given the surroundings, might have been the most fun thing he could think of. If you're autistic, then focusing obsessively on some task might be a great way to pass the time, but if you're more or less normal, I doubt you'll get very far with obsessive-compulsive self-training -- and that's the problem, isn't it? If the signals have properties of their own, I'm afraid they will start interfering with each other, which won't allow the circuit to execute in real time. Binary signals, on the other hand, can be encoded by the activation of nodes of the circuit, active/inactive. If you have an AND gate that leads from symbols S1 and S2 to S3, you learn to remember S3 only when you see both S1 and S2 (probably you'll still need complementary symbol to develop negative, so you'll also need -S1, -S2 and -S3, so that -S3 is activated (recalled) when you see S1 and -S2, whole table. You'll also need separate symbols for each node in each gate. Probably randomly generated hieroglyph-like symbols are a good way to create new categories in the mind for new nodes in the circuit, and also to train yourself to recall the right answers on the gates, by drawing them together. -- Vladimir Nesov [EMAIL PROTECTED] http://causalityrelay.wordpress.com/ --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
RE: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
be modified to perform a lot of tasks for which many people in AI still think there is no method for solving, such as complex context-appropriate inferencing. Hinton's paper shows that neural net learning is suddenly much more powerful than it has been before. And Hecht-Neilsen's paper shows another powerful form or neural net-like learning and computing that scales well. The convergence of such much more sophisticated software approaches and the much more powerful hardware necessary to actually build minds that use them is much more than just a belief. Today, for $33K you can buy a system I talked about in my email which started this thread. It has 126Gbytes of RAM and roughly 160Million random RAM access/second. This is enough power to start building small toy AGI mind that could show limited generalized learning, perception, inferencing, planning, behaviors, attention focusing, and behavior selection, i.e., something like Ben's pet brains. The $850K system would allow substantially more sophisticated demonstrations of artificial minds to be created. This combination of much more sophisticated understandings for how to build AGI's, combined with much more powerful hardware is something new. And, much, much more powerful hardware should be arriving in about 6 years when multi-level chips with mesh-networked, massively-mutli-cored processors, and 8 or more layers of memory connected to the processors with many thousands of though silicon vias, and with hundreds of high speed channels to external memory and other such multi-level chips will hopefully become routinely available. Richard, a lot has changed since the '70, '80, '90s, and early '00s --- and if you do see it --- that's your problem. Ed Porter -Original Message- From: Richard Loosemore [mailto:[EMAIL PROTECTED] Sent: Saturday, June 28, 2008 4:14 PM To: agi@v2.listbox.com Subject: Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI Ed Porter wrote: I do not claim the software architecture for AGI has been totally solved. But I believe that enough good AGI approaches exist (and I think Novamente is one) that when powerful hardware available to more people we will be able to relatively quickly get systems up and running that demonstrate the parts of the problems we have solved. And that will provide valuable insights and test beds for solving the parts of the problem that we have not yet solved. You are not getting my point. What you just said was EXACTLY what was said in 1970, 1971, 1972, 1973 ..2003, 2004, 2005, 2006, 2007 .. And every time it was said, the same justification for the claim was given: I just have this belief that it will work. Plus ca change, plus c'est la meme fubar. With regard to your statement the problem is understanding HOW TO DO IT --- WE DO UNDERSTAND HOW TO DO IT --- NOT ALL OF IT --- AND NOT HOW TO MAKE IT ALL WORK TOGETHER WELL AUTOMATICALLY --- BUT --- GIVEN THE TYPE OF HARDWARE EXPECTED TO COST LESS THAN $3M IN 6 YEARS --- WE KNOW HOW TO BUILD MUCH OF IT --- ENOUGH THAT WE COULD PROVIDE EXTREMELY VALUABLE COMPUTERS WITH OUR CURRENT UNDERSTANDINGS. You do *not* understand how to do it. But I have to say that statements like your paragraph above are actually very good for my health, because their humor content is right up there in the top ten, along with Eddie Izzard's Death Star Canteen sketch and Stephen Colbert at the 2006 White House Correspondents' Association Dinner. So long as the general response to the complex systems problem is not This could be a serious issue, let's put our heads together to investigate it, but My gut feeling is that this is just not going to be a problem, or Quit rocking the boat!, you can bet that nobody really wants to ask any questions about whether the approaches are correct, they just want to be left alone to get on with their approaches. History, I think, will have some interesting things to say about all this. Good luck anyway. Richard Loosemore --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Ed:Another reason for optimism is Hintons new work described in papers such as Modeling image patches with a directed hierarchy of Markov random fields by Simon Osindero and Geoffrey Hinton and the Google Tech Talk at http://www.youtube.com/watch?v=AyzOUbkUf3M. Hinton has shown how to automatically learn hierarchical neural nets that have 2000 hidden nodes in one layer, 500 in the next, and 1000 in the top layer Comment from a pal on Hinton who was similarly recommended on slashdot:(I'm ignorant here): I also took a closer look at the Hinton stuff that the slashdot poster made reference to. To call this DBN stuff highly advanced over Hawkins is ridiculous. I looked at it already a couple of months ago. It took Hinton ***17-years*** - by his own admission - to figure out how to build a connectionist net that could reliably identify variations of handwritten numbers 1-9. And it's gonna take him about a MILLION more years to do general AI with this approach. Gakk. To me, the biggest problem with connectionist networks is all they ever solve are toy problems - and it's 20 years after connectionism become popular again. --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
RE: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Mike, None of the Hawkin's papers I have read have given any results as impressive as the Hinton papers I cited. If you know of some that have, please send me references to the most impressive among them. Hinton says he believe his system could scale efficiently to much larger nets. If that is true, a system having multiples of his modules would appear possibly able to learn how to handle a good chunk of sensory perception. Like, Ben I am not wed to a totally connectionist approach, but rather one that has attributes of both connectionist and symbolic approaches. I personally like to think in terms of systems where I have some idea what things represent, so I can think in terms of what I want them to do. But still I am impressed with what Hinton has shown, particularly if it can be made to scale well to much larger systems. Ed Porter -Original Message- From: Mike Tintner [mailto:[EMAIL PROTECTED] Sent: Sunday, June 29, 2008 2:48 PM To: agi@v2.listbox.com Cc: [EMAIL PROTECTED] Subject: Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI Ed:Another reason for optimism is Hintons new work described in papers such as Modeling image patches with a directed hierarchy of Markov random fields by Simon Osindero and Geoffrey Hinton and the Google Tech Talk at http://www.youtube.com/watch?v=AyzOUbkUf3M. Hinton has shown how to automatically learn hierarchical neural nets that have 2000 hidden nodes in one layer, 500 in the next, and 1000 in the top layer Comment from a pal on Hinton who was similarly recommended on slashdot:(I'm ignorant here): I also took a closer look at the Hinton stuff that the slashdot poster made reference to. To call this DBN stuff highly advanced over Hawkins is ridiculous. I looked at it already a couple of months ago. It took Hinton ***17-years*** - by his own admission - to figure out how to build a connectionist net that could reliably identify variations of handwritten numbers 1-9. And it's gonna take him about a MILLION more years to do general AI with this approach. Gakk. To me, the biggest problem with connectionist networks is all they ever solve are toy problems - and it's 20 years after connectionism become popular again. --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
RE: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
I agree that the hardware advances are inspirational, and it seems possible that just having huge hardware around could change the way people think and encourage new ideas. But what I'm really looking forward to is somebody producing a very impressive general intelligence result that was just really annoying because it took 10 days of computing instead of an hour. Seems to me that all the known AGI researchers are in theory, design, or system building phases; I don't think any of them are CPU-bound at present -- and no fair pointing to Goedel Machines or AIXI either, which will ALWAYS be resource-starved :) --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Ben Goertzel wrote: Richard, So long as the general response to the complex systems problem is not This could be a serious issue, let's put our heads together to investigate it, but My gut feeling is that this is just not going to be a problem, or Quit rocking the boat!, you can bet that nobody really wants to ask any questions about whether the approaches are correct, they just want to be left alone to get on with their approaches. Both Ed Porter and myself have given serious thought to the complex systems problem as you call it, and have discussed it with you at length. I also read the only formal paper you sent me dealing with it (albeit somewhat indirectly) and also your various online discourses on the topic. Ed and I don't agree with you on the topic, but not because of lack of thinking or attention. Your argument FOR the existence of a complex systems problem with Novamente or OpenCog, is not any more rigorous than our argument AGAINST it. Oh, mere rhetoric. You have never given an argument against it. If you believe this is not correct, perhaps you could jog my memory by giving a brief summary of what you think is the argument against it? In all of my discussions with you on the subject, you have introduced many red herrings, and we have discussed many topics that turned out to be just misunderstandings, but you have never addressed the actual core argument itself. In fact, IIRC, on the one occasion that I persisted in trying to bring the discussion back to the core issue, you finally made only one argument against my core claim your argument against it was I just don't think it is going to be a problem. The argument itself is extremely rigorous: on all the occasions on which someone has disputed the rigorousness of the argument, they have either addressed some other issue entirely or they have just waved their hands without showing any sign of understanding the argument, and then said ... it's not rigorous!. It is almost comical to go back over the various responses to the argument: not only do people go flying off in all sorts of bizarre directions, but they also get quite strenuous about it at the same time. Not understanding an argument is not the same as the argument not being rigorous. Richard Loosemore --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
On Friday 27 June 2008, Richard Loosemore wrote: Pardon my fury, but the problem is understanding HOW TO DO IT, and HOW TO BUILD THE TOOLS TO DO IT, not having expensive hardware. So long as some people on this list repeat this mistake, this list will degenerate even further into obsolescence. I am working on this issue, but it will not look like ai from your perspective. It is, in a sense, ai. Here's the tool approach: http://heybryan.org/buildingbrains.html http://heybryan.org/exp.html Sort of. - Bryan http://heybryan.org/ --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
The argument itself is extremely rigorous: on all the occasions on which someone has disputed the rigorousness of the argument, they have either addressed some other issue entirely or they have just waved their hands without showing any sign of understanding the argument, and then said ... it's not rigorous!. It is almost comical to go back over the various responses to the argument: not only do people go flying off in all sorts of bizarre directions, but they also get quite strenuous about it at the same time. Richard, if your argument is so rigorous, why don't you do this: present a brief, mathematical formalization of your argument, defining all terms precisely and carrying out all inference steps exactly, at the level of a textbook mathematical proof. I'll be on vacation for the next 2 weeks w/limited and infrequent email access, so I'll look out for this when I return. If you present your argument this way, then you can rest assured I will understand it, as I'm capable to understand math; then, our arguments can be more neatly directed ... toward the appropriateness of your formal definitions and assumptions... -- Ben G --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
virtually impossible to train a neural net with so many hidden nodes, but Hinton's new method allows rapid largely automatic training of such large networks, enabling in the example show, surprisingly good handwritten numeral recognition. Don't have access to that paper right now, so can you tell me: this goes beyond mere supervised learning, right? And it solves the problem of representing multiple tokens? And also the problem of encoding structured knowledge? It doesn't represent structure with hard-coded templates, yes? The technique scales well to full-scale thinking systems in which the domain is not restricted to, say, handwriting, but includes everything the system could ever want to recognize, yes? Oh. and in case I forget, the images are not preselected, but are naturally occuring in context, so the system can recognize a letter A in a scene in which two people lean against one another and hold something horizontal between them at waist height? I assume all the answers to the above were Yes!, so it sounds like a great leap forward: I'll read the full paper tomorrow. Pity that Hinton chose a title that implied all the answers were 'no'. Bit of an oversight on his part, but never mind. Yet another example of the power of automatic learning is shown by impressive success of Hecht-Nielsen confabulation system in generating a second sentence that reasonably follows from first, as if it had been written by a human intelligence, withoug any attempt to teach the rules of grammar or any explicit semantic knowledge. The system learns from text corpora. You may say this is narrow AI. But it all has general applicability. For example, the type of hierarchical memory with max-pooling shown in Serre's paper shows is an extremely powerful paradigm that addresses some of the most difficult problems in AI, including robust non-literal matching. Such hierarchical memory can be modified to perform a lot of tasks for which many people in AI still think there is no method for solving, such as complex context-appropriate inferencing. Hinton's paper shows that neural net learning is suddenly much more powerful than it has been before. And Hecht-Neilsen's paper shows another powerful form or neural net-like learning and computing that scales well The convergence of such much more sophisticated software approaches and the much more powerful hardware necessary to actually build minds that use them is much more than just a belief. Today, for $33K you can buy a system I talked about in my email which started this thread. It has 126Gbytes of RAM and roughly 160Million random RAM access/second. This is enough power to start building small toy AGI mind that could show limited generalized learning, perception, inferencing, planning, behaviors, attention focusing, and behavior selection, i.e., something like Ben's pet brains. The $850K system would allow substantially more sophisticated demonstrations of artificial minds to be created. This combination of much more sophisticated understandings for how to build AGI's, combined with much more powerful hardware is something new. And, much, much more powerful hardware should be arriving in about 6 years when multi-level chips with mesh-networked, massively-mutli-cored processors, and 8 or more layers of memory connected to the processors with many thousands of though silicon vias, and with hundreds of high speed channels to external memory and other such multi-level chips will hopefully become routinely available. Richard, a lot has changed since the '70, '80, '90s, and early '00s --- and if you do see it --- that's your problem. Oh dear, Ed. I just shouldn't get into discussions with you. It's fun sometimes, but. Back to work. Richard Loosemore Ed Porter -Original Message- From: Richard Loosemore [mailto:[EMAIL PROTECTED] Sent: Saturday, June 28, 2008 4:14 PM To: agi@v2.listbox.com Subject: Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI Ed Porter wrote: I do not claim the software architecture for AGI has been totally solved. But I believe that enough good AGI approaches exist (and I think Novamente is one) that when powerful hardware available to more people we will be able to relatively quickly get systems up and running that demonstrate the parts of the problems we have solved. And that will provide valuable insights and test beds for solving the parts of the problem that we have not yet solved. You are not getting my point. What you just said was EXACTLY what was said in 1970, 1971, 1972, 1973 ..2003, 2004, 2005, 2006, 2007 .. And every time it was said, the same justification for the claim was given: I just have this belief that it will work. Plus ca change, plus c'est la meme fubar. With regard to your statement the problem is understanding HOW TO DO IT --- WE DO UNDERSTAND HOW TO DO IT --- NOT ALL OF IT --- AND NOT HOW TO MAKE IT ALL WORK
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Ben Goertzel wrote: The argument itself is extremely rigorous: on all the occasions on which someone has disputed the rigorousness of the argument, they have either addressed some other issue entirely or they have just waved their hands without showing any sign of understanding the argument, and then said ... it's not rigorous!. It is almost comical to go back over the various responses to the argument: not only do people go flying off in all sorts of bizarre directions, but they also get quite strenuous about it at the same time. Richard, if your argument is so rigorous, why don't you do this: present a brief, mathematical formalization of your argument, defining all terms precisely and carrying out all inference steps exactly, at the level of a textbook mathematical proof. I'll be on vacation for the next 2 weeks w/limited and infrequent email access, so I'll look out for this when I return. If you present your argument this way, then you can rest assured I will understand it, as I'm capable to understand math; then, our arguments can be more neatly directed ... toward the appropriateness of your formal definitions and assumptions... Mathematics is about formal systems. The argument is not about formal systems, it is about real-world intelligent systems and their limitations, and about the very *question* of whether those intelligent systems are formal systems. It is about whether scientific methodology (which is just the exercise of a particular subset of this thing we call 'intelligence') is itself a formal system. To formulate the argument in mathematical terms would, therefore, be to prejudge the answer to the question we are addressing - nothing could more silly than to insist on a mathematical formulation of it. Asking for a mathematical formulation of an argument that has nothing to do with formal systems is, therefore, a sign that you have no understanding of what the argument is actually about. Now, if it were anyone else I would say that you really did not understand, and were just, well ignorant. But you actually do understand that point: when you made the above request I think your goal was to engage in a piece of pure sophistry. You cynically ask for something that you know has no relevance, and cannot be supplied, as an attempt at a put-down. Nice try, Ben. Or, then again . perhaps I am wrong: maybe you really *cannot* understand anything except math? Perhaps you have no idea what the actual argument is, and that has been the problem all along? I notice that you avoided answering my request that you summarize your argument against the complex systems problem ... perhaps you are just confused about what the argument actually is, and have been confused right from the beginning? Richard Loosemore --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Richard, I think that it would be possible to formalize your complex systems argument mathematically, but I don't have time to do so right now. Or, then again . perhaps I am wrong: maybe you really *cannot* understand anything except math? It's not the case that I can only understand math -- however, I have a lot of respect for the power of math to clarify disagreements. Without math, arguments often proceed in a confused way because different people are defining terms differently,a and don't realize it. But, I agree math is not the only kind of rigor. I would be happy with a very careful, systematic exposition of your argument along the lines of Spinoza or the early Wittgenstein. Their arguments were not mathematical, but were very rigorous and precisely drawn -- not slippery. Perhaps you have no idea what the actual argument is, and that has been the problem all along? I notice that you avoided answering my request that you summarize your argument against the complex systems problem ... perhaps you are just confused about what the argument actually is, and have been confused right from the beginning? In a nutshell, it seems you are arguing that general intelligence is fundamentally founded on emergent properties of complex systems, and that it's not possible for us to figure out analytically how these emergent properties emerge from the lower-level structures and dynamics of the complex systems involved. Evolution, you suggest, figured out some complex systems that give rise to the appropriate emergent properties to produce general intelligence. But evolution did not do this figuring-out in an analytical way, rather via its own special sort of directed trial and error. You suggest that to create a generally intelligent system, we should create a software framework that makes it very easy to experiment with different sorts of complex systems, so that we can then figure out (via some combination of experiment, analysis, intuition, theory, etc.) how to create a complex system that gives rise to the emergent properties associated with general intelligence. I'm sure the above is not exactly how you'd phrase your argument -- and it doesn't capture all the nuances -- but I was trying to give a compact and approximate formulation. If you'd like to give an alternative, equally compact formulation, that would be great. I think the flaw of your argument lies in your definition of complexity, and that this would be revealed if you formalized your argument more fully. I think you define complexity as a kind of fundamental irreducibility that the human brain does not possess, and that engineered AGI systems need not possess. I think that real systems display complexity which makes it **computationally difficult** to explain their emergent properties in terms of their lower-level structures and dynamics, but not as fundamentally intractable as you presume. But because you don't formalize your notion of complexity adequately, it's not possible to engage you in rational argumentation regarding the deep flaw at the center of your argument. However, I cannot prove rigorously that the brain is NOT complex in the overly strong sense you allude it is ... and nor can I prove rigorously that a design like Novamente Cognition Engine or OpenCog Prime will give rise to the emergent properties associated with general intelligence. So, in this sense, I don't have a rigorous refutation of your argument, and nor would I if you rigorously formalized your argument. However, I think a rigorous formulation of your argument would make it apparent to nearly everyone reading it that your definition of complexity is unreasonably strong. -- Ben G --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
RE: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
on massively parallel systems efficiently. If anything, the problem right now is the confusion of possible approaches to many of the problems. More cheap hardware will allow more of them to be tested of systems of the necessary complexity, and the better ones to become more widely accepted RICHARD LOOSEMORE Frankly, looking at recent posts, I think this list is already dead. ED PORTER Richard, if the list is so dead of late, how come you have posted to it so often recently? -Original Message- From: Richard Loosemore [mailto:[EMAIL PROTECTED] Sent: Friday, June 27, 2008 4:30 PM To: agi@v2.listbox.com Subject: Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI At a quick glance I would say you could do it cheaper by building it yourself rather than buying Dell servers (cf MicroWulf project that was discussed before: http://www.clustermonkey.net//content/view/211/33/). Secondly: if what you need to get done is spreading activation (which implies massive parallelism) you would probably be better off with a Celoxica system than COTS servers: celoxica.com. Hugo de Garis has a good deal of experience with using this hardware: it is FPGA based, so the potential parallelism is huge. Third: the problem, in any case, is not the hardware. AI researchers have saying if only we had better hardware, we could really get these algorithms to sing, and THEN we will have a real AI! since the f***ing 1970s, at least. There is nothing on this earth more stupid than watching people repeat the same mistakes over and over again, for decades in a row. Pardon my fury, but the problem is understanding HOW TO DO IT, and HOW TO BUILD THE TOOLS TO DO IT, not having expensive hardware. So long as some people on this list repeat this mistake, this list will degenerate even further into obsolescence. Frankly, looking at recent posts, I think this list is already dead. Richard Loosemore Ed Porter wrote: WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI On Wednesday, June 25, US East Cost time, I had an interesting phone conversation with Dave Hart, where we discussed just how much hardware could you get for the current buck, for the amounts of money AGI research teams using OpenCog (THE LUCKY ONES) might have available to them. After our talk I checked out the cost of current servers at Dell (the easiest place I knew of to check out prices.) I found hardware, and particularly memory was somewhat cheaper than Dave and I had thought. But it is still sufficiently expensive, that moderately funded projects are going to be greatly limited by the processor-memory and inter-processor bandwidth as to how much spreading activation and inferencing they will be capable of doing. A RACK MOUNTABLE SERVER WITH 4 QUAD-CORE XEONS, WITH EACH PROCESSOR HAVING 8MB OF CACHE, AND THE WHOLE SERVER HAVING 128GBYTES OF RAM AND FOUR 300GBYTE HARD DRIVES WAS UNDER $30K. The memory stayed roughly constant in price per GByte going from 32 to 64 to 128 GBytes. Of course you would probably have to pay a several extra grand for software and warranties. SO LET US SAY THE PRICE IS $33K PER SERVER. A 24 port 20Gbit/sec infiniband switch with cables and one 20Gbit/sec adapter card for each of 24 servers would be about $52K SO A TOTAL SYSTEM WITH 24 SERVERS, 96 PROCESSORS, 384 CORES, 768MBYTE OF L2 CACHE, 3 TBYTES OF RAM, AND 28.8TBYTES OF DISK, AND THE 24 PORT 20GBIT/SEC SWITCH WOULD BE ROUGHLY $850 GRAND. That doesn't include air conditioning. I am guessing each server probably draws about 400 watts, so 24 of them would be about 9600 watts--- about the amount of heat of ten hair dryers running in one room, which obviously would require some cooling, but I would not think would be that expensive to handle. With regard to performance, such systems are not even close to human brain level but they should allow some interesting proofs of concepts Performance --- AI spreading activation often involves a fair amount of non-locality of memory. Unfortunately there is a real penalty for accessing RAM randomly. Without overleaving, one article I read recently implied about 50ns was a short latency for a memory access. So we will assume 20M random RAM access (randomRamOpps) per second per channel, and that an average activation will take two, a read and write, so roughly 10M activations/sec per memory channel. Matt Mahoney has pointed out that spreading activation can be modeled by matrix methods that let you access RAM with much higher sequential memory accessing rates. He claimed he could process about a gigabyte of matrix data a second. If one assumes each element in the matrix is 8 bytes, that would be the equivalent of doing 125M activation a second, which is roughly 12.5 times faster (if just 2 bytes, it would be 50 times fasters, or 500M activation/sec). If one
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
There was one little line in this post that struck me, and I wanted to comment: Quoting Ed Porter [EMAIL PROTECTED]: With regard to performance, such systems are not even close to human brain level but they should allow some interesting proofs of concepts Mentioning some huge system. My thought was, wow, that's just sounds sad. But I guess it depends on what you mean by performance. One thing that computers now way exceed brain performance in is reliability of the operations. Sure, it's difficult to say what a basic brain operation is (is a synapse reaction equivalent to a multiply accumulate?), but one thing that can be said about them is that they aren't very reliable or precise. They have a sort of a range of operation, where they kind of will act in a certain way given an input. It's got to be really hard to get valuable behavior out of this kind of a system, so the brain uses massive redundancy. Now, it might well be that in addition to just the reliability, this kind of a system gets other value from it, like a nice probabilistic operation that has additional value in itself. Maybe the inherent unpredictability is part of what we mean by intelligence. Personally I suspect that to be true. But this all stands in great contrast to how computers naturally work--obeying information processing instructions with absolute precision (possibly error-free, depending on how you look it). There is a sort of mismatch between good human brain behavior and good computer behavior. It seems like the AGI project is about making a computer act like a good brain. We can focus on how to get a computer to act in ways that are ideal for a brain to act intelligently. And by this I mean something like having some basic operations and systems that can be used in all situations. But I think it might also be good to try to think of it in terms of looking for the best ways for a computer to be intelligent. I'm a patchwork AGI kind of guy, and while surely there must be some general mechanism, it seems to make sense that there could also be many very finely crafted modules. Unfortunately, if we are restricting modules to human written modules, then that's the basic problem. A basic function of an AGI should be that it can write programs for itself to handle tasks. Or I guess for other systems. But if it can do that, then these programs don't need such huge amounts of computer power. andi --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Ed Porter wrote: I do not claim the software architecture for AGI has been totally solved. But I believe that enough good AGI approaches exist (and I think Novamente is one) that when powerful hardware available to more people we will be able to relatively quickly get systems up and running that demonstrate the parts of the problems we have solved. And that will provide valuable insights and test beds for solving the parts of the problem that we have not yet solved. You are not getting my point. What you just said was EXACTLY what was said in 1970, 1971, 1972, 1973 ..2003, 2004, 2005, 2006, 2007 .. And every time it was said, the same justification for the claim was given: I just have this belief that it will work. Plus ca change, plus c'est la meme fubar. With regard to your statement the problem is understanding HOW TO DO IT --- WE DO UNDERSTAND HOW TO DO IT --- NOT ALL OF IT --- AND NOT HOW TO MAKE IT ALL WORK TOGETHER WELL AUTOMATICALLY --- BUT --- GIVEN THE TYPE OF HARDWARE EXPECTED TO COST LESS THAN $3M IN 6 YEARS --- WE KNOW HOW TO BUILD MUCH OF IT --- ENOUGH THAT WE COULD PROVIDE EXTREMELY VALUABLE COMPUTERS WITH OUR CURRENT UNDERSTANDINGS. You do *not* understand how to do it. But I have to say that statements like your paragraph above are actually very good for my health, because their humor content is right up there in the top ten, along with Eddie Izzard's Death Star Canteen sketch and Stephen Colbert at the 2006 White House Correspondents' Association Dinner. So long as the general response to the complex systems problem is not This could be a serious issue, let's put our heads together to investigate it, but My gut feeling is that this is just not going to be a problem, or Quit rocking the boat!, you can bet that nobody really wants to ask any questions about whether the approaches are correct, they just want to be left alone to get on with their approaches. History, I think, will have some interesting things to say about all this. Good luck anyway. Richard Loosemore --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
RE: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Wannabe, Your qualification is totally appropriate. When I say such systems are not even close to human brain level I mean not close to human level at the types of things human brains current outperform computers. Obviously there are many ways in which just current PC's outperform human by thousands or millions of times. But there still are many tasks at which human minds greatly outperform computers, and that is where a lot of the focus in AGI is. By a human level AGI, I mean a computer that can do almost all the things a human brain does as fast as a human. But such hardware will probably be capable of performing many of the things a PC can already do much faster than a human, many times faster than a PC. A machine that can do all the types of things a human does as fast as a human, and that can also do many tasks millions of times faster than a human --- and that can mix and match, blend, and interface between these two different types of processes rapidly will be extremely powerful. For example, such a system could scan text at very high speeds (millions of pages a second), and where it found combinations of words that looked interesting slow down and read them at a fast skim (10s to 1000s of times faster than a human), and then read the texts from the skim that seem interesting at roughly human speed would be able to find and understand relevant information in large volumes of thousands of times faster than a human. And of course, once such texts have been read they would be indexed and be much more rapidly available for future access when relevant. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Saturday, June 28, 2008 3:36 PM To: agi@v2.listbox.com Subject: Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI There was one little line in this post that struck me, and I wanted to comment: Quoting Ed Porter [EMAIL PROTECTED]: With regard to performance, such systems are not even close to human brain level but they should allow some interesting proofs of concepts Mentioning some huge system. My thought was, wow, that's just sounds sad. But I guess it depends on what you mean by performance. One thing that computers now way exceed brain performance in is reliability of the operations. Sure, it's difficult to say what a basic brain operation is (is a synapse reaction equivalent to a multiply accumulate?), but one thing that can be said about them is that they aren't very reliable or precise. They have a sort of a range of operation, where they kind of will act in a certain way given an input. It's got to be really hard to get valuable behavior out of this kind of a system, so the brain uses massive redundancy. Now, it might well be that in addition to just the reliability, this kind of a system gets other value from it, like a nice probabilistic operation that has additional value in itself. Maybe the inherent unpredictability is part of what we mean by intelligence. Personally I suspect that to be true. But this all stands in great contrast to how computers naturally work--obeying information processing instructions with absolute precision (possibly error-free, depending on how you look it). There is a sort of mismatch between good human brain behavior and good computer behavior. It seems like the AGI project is about making a computer act like a good brain. We can focus on how to get a computer to act in ways that are ideal for a brain to act intelligently. And by this I mean something like having some basic operations and systems that can be used in all situations. But I think it might also be good to try to think of it in terms of looking for the best ways for a computer to be intelligent. I'm a patchwork AGI kind of guy, and while surely there must be some general mechanism, it seems to make sense that there could also be many very finely crafted modules. Unfortunately, if we are restricting modules to human written modules, then that's the basic problem. A basic function of an AGI should be that it can write programs for itself to handle tasks. Or I guess for other systems. But if it can do that, then these programs don't need such huge amounts of computer power. andi --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
On Sat, Jun 28, 2008 at 4:13 PM, Richard Loosemore [EMAIL PROTECTED] wrote: Ed Porter wrote: I do not claim the software architecture for AGI has been totally solved. But I believe that enough good AGI approaches exist (and I think Novamente is one) that when powerful hardware available to more people we will be able to relatively quickly get systems up and running that demonstrate the parts of the problems we have solved. And that will provide valuable insights and test beds for solving the parts of the problem that we have not yet solved. You are not getting my point. What you just said was EXACTLY what was said in 1970, 1971, 1972, 1973 ..2003, 2004, 2005, 2006, 2007 .. And every time it was said, the same justification for the claim was given: I just have this belief that it will work. It is not the case that the reason I believe Novamente/OpenCog can work for AGI is just a belief Nor, however, is the reason an argument that can be summarized in an email. I'm setting out on a 2-week vacation on Monday (June 30 - July 13), on which I'll be pretty much without email (in the wilds of Alaska ;-) ... so it's a bad time for me to get involved in deep discussions But I hope to release some docs on OpenCog Prime later this summer, which will disclose a bit more of my reasons for thinking the approach can succeed. Ed has seen much of this material before, but most others on this list have not... There is a broad range of qualities-of-justification, between a mere belief on the one hand, and a rigorous proof on the other. -- Ben G --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Richard and Ed, Insanity is doing the same thing over and over again and expecting different results. - Albert Einstein Prelude to insanity: unintentionally doing the same thing over and over again and getting the same results. - Me Cheers, Brad Richard Loosemore wrote: Ed Porter wrote: I do not claim the software architecture for AGI has been totally solved. But I believe that enough good AGI approaches exist (and I think Novamente is one) that when powerful hardware available to more people we will be able to relatively quickly get systems up and running that demonstrate the parts of the problems we have solved. And that will provide valuable insights and test beds for solving the parts of the problem that we have not yet solved. You are not getting my point. What you just said was EXACTLY what was said in 1970, 1971, 1972, 1973 ..2003, 2004, 2005, 2006, 2007 .. And every time it was said, the same justification for the claim was given: I just have this belief that it will work. Plus ca change, plus c'est la meme fubar. With regard to your statement the problem is understanding HOW TO DO IT --- WE DO UNDERSTAND HOW TO DO IT --- NOT ALL OF IT --- AND NOT HOW TO MAKE IT ALL WORK TOGETHER WELL AUTOMATICALLY --- BUT --- GIVEN THE TYPE OF HARDWARE EXPECTED TO COST LESS THAN $3M IN 6 YEARS --- WE KNOW HOW TO BUILD MUCH OF IT --- ENOUGH THAT WE COULD PROVIDE EXTREMELY VALUABLE COMPUTERS WITH OUR CURRENT UNDERSTANDINGS. You do *not* understand how to do it. But I have to say that statements like your paragraph above are actually very good for my health, because their humor content is right up there in the top ten, along with Eddie Izzard's Death Star Canteen sketch and Stephen Colbert at the 2006 White House Correspondents' Association Dinner. So long as the general response to the complex systems problem is not This could be a serious issue, let's put our heads together to investigate it, but My gut feeling is that this is just not going to be a problem, or Quit rocking the boat!, you can bet that nobody really wants to ask any questions about whether the approaches are correct, they just want to be left alone to get on with their approaches. History, I think, will have some interesting things to say about all this. Good luck anyway. Richard Loosemore --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
Richard, So long as the general response to the complex systems problem is not This could be a serious issue, let's put our heads together to investigate it, but My gut feeling is that this is just not going to be a problem, or Quit rocking the boat!, you can bet that nobody really wants to ask any questions about whether the approaches are correct, they just want to be left alone to get on with their approaches. Both Ed Porter and myself have given serious thought to the complex systems problem as you call it, and have discussed it with you at length. I also read the only formal paper you sent me dealing with it (albeit somewhat indirectly) and also your various online discourses on the topic. Ed and I don't agree with you on the topic, but not because of lack of thinking or attention. Your argument FOR the existence of a complex systems problem with Novamente or OpenCog, is not any more rigorous than our argument AGAINST it. Similarly, I have no rigorous argument that Novamente and OpenCog won't fail because of the lack of a soul. I can't prove this formally -- and even if I did, those who believe a soul is necessary for AI could always dispute the mathematical assumptions of my proof. And those who do claim a soul is necessary, have no rigorous arguments in their favor, except ones based transparently on assumptions I reject... And so it goes... Ben --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=106510220-47b225 Powered by Listbox: http://www.listbox.com
Re: [agi] WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI
At a quick glance I would say you could do it cheaper by building it yourself rather than buying Dell servers (cf MicroWulf project that was discussed before: http://www.clustermonkey.net//content/view/211/33/). Secondly: if what you need to get done is spreading activation (which implies massive parallelism) you would probably be better off with a Celoxica system than COTS servers: celoxica.com. Hugo de Garis has a good deal of experience with using this hardware: it is FPGA based, so the potential parallelism is huge. Third: the problem, in any case, is not the hardware. AI researchers have saying if only we had better hardware, we could really get these algorithms to sing, and THEN we will have a real AI! since the f***ing 1970s, at least. There is nothing on this earth more stupid than watching people repeat the same mistakes over and over again, for decades in a row. Pardon my fury, but the problem is understanding HOW TO DO IT, and HOW TO BUILD THE TOOLS TO DO IT, not having expensive hardware. So long as some people on this list repeat this mistake, this list will degenerate even further into obsolescence. Frankly, looking at recent posts, I think this list is already dead. Richard Loosemore Ed Porter wrote: WHAT SORT OF HARDWARE $33K AND $850K BUYS TODAY FOR USE IN AGI On Wednesday, June 25, US East Cost time, I had an interesting phone conversation with Dave Hart, where we discussed just how much hardware could you get for the current buck, for the amounts of money AGI research teams using OpenCog (THE LUCKY ONES) might have available to them. After our talk I checked out the cost of current servers at Dell (the easiest place I knew of to check out prices.) I found hardware, and particularly memory was somewhat cheaper than Dave and I had thought. But it is still sufficiently expensive, that moderately funded projects are going to be greatly limited by the processor-memory and inter-processor bandwidth as to how much spreading activation and inferencing they will be capable of doing. A RACK MOUNTABLE SERVER WITH 4 QUAD-CORE XEONS, WITH EACH PROCESSOR HAVING 8MB OF CACHE, AND THE WHOLE SERVER HAVING 128GBYTES OF RAM AND FOUR 300GBYTE HARD DRIVES WAS UNDER $30K. The memory stayed roughly constant in price per GByte going from 32 to 64 to 128 GBytes. Of course you would probably have to pay a several extra grand for software and warranties. SO LET US SAY THE PRICE IS $33K PER SERVER. A 24 port 20Gbit/sec infiniband switch with cables and one 20Gbit/sec adapter card for each of 24 servers would be about $52K SO A TOTAL SYSTEM WITH 24 SERVERS, 96 PROCESSORS, 384 CORES, 768MBYTE OF L2 CACHE, 3 TBYTES OF RAM, AND 28.8TBYTES OF DISK, AND THE 24 PORT 20GBIT/SEC SWITCH WOULD BE ROUGHLY $850 GRAND. That doesn't include air conditioning. I am guessing each server probably draws about 400 watts, so 24 of them would be about 9600 watts--- about the amount of heat of ten hair dryers running in one room, which obviously would require some cooling, but I would not think would be that expensive to handle. With regard to performance, such systems are not even close to human brain level but they should allow some interesting proofs of concepts Performance --- AI spreading activation often involves a fair amount of non-locality of memory. Unfortunately there is a real penalty for accessing RAM randomly. Without overleaving, one article I read recently implied about 50ns was a short latency for a memory access. So we will assume 20M random RAM access (randomRamOpps) per second per channel, and that an average activation will take two, a read and write, so roughly 10M activations/sec per memory channel. Matt Mahoney has pointed out that spreading activation can be modeled by matrix methods that let you access RAM with much higher sequential memory accessing rates. He claimed he could process about a gigabyte of matrix data a second. If one assumes each element in the matrix is 8 bytes, that would be the equivalent of doing 125M activation a second, which is roughly 12.5 times faster (if just 2 bytes, it would be 50 times fasters, or 500M activation/sec). If one assumes each of 4 core of each of 4 processors could handle a matrix at 1GByte/sec, and each element in the matrix was just 2 bytes, that would be 8 G 2Byte matrix activations/sec/server, and 256G matrix activation/sec/system. It is not clear how well this could be made to work with the type of interconnectivity of an AGI. It is clear their would be some penalty for sparseness, perhaps a large one. If one used run-length encoding in matrix, which is read by rows, then a set of column whose values could fit in cache could be loaded into cache, and the portions of all the rows relating to them could be read sequentially. Once all the portions of all the row relating to the sub-set of colums had been processed, then the process could be repeated for another